1

�����������

�������

��� �� ������

����������������� ���������� ��

������� �!"�#$����������

%��� &���������� ��%� ����

'%��������� ���(������������ ����� ������������������������������������

� ������� �������

������� ��!"�� ##$%&$'(��$% $)*

+���������"����� , �!����%�����������������+-��(����� !����.�&�+-���/��(�&���� !�������0/��

��� 0!�%���������� ���

�/1��!������ &��23%#$���

4�������1��"� �567����������/���/��!���(�.������/��!��#�(�#�-����������������������������������

6������ �(�������'89���!� ���� ���������

4��!�����"������ '����!/������.��!! ���/�������/�0���0"�!/���/��/���.!��

)%������&�� ���(��!�1����������"� �������/������:)-;9<+-=%>7��!/��/������� ��

7/��!�/���������� >/� /!����������.�������/���

4/�/���"������"�!�� &�0/��������"�!�

:�����/� ����.� ���� 9��%������������������&$%&&?4�

)%������&�� ��� ���(+������������� �������!������%��������#�����!�������� !��

:�����/� ����������� &��"���������.������������/� �����������/���0/��� �./���/���.���/�0�/���

������/� �����������!! ���/�������/�0���

9!/���� <!/������!����.�������/�������/� �(��� ���/!/���.���������

�� ������ ��� ��*�������+����,�(7/��������/����� 7/��@��/���0/���(����� �������(�����/���(�����%!/0�!!����0/������������!

;��������������� 4"�!���!! ���/���������������1�����������"(���������"�/���������/� ����������

4"�!��������� 5������$$$��!! ���/������.�!�����������������������/!������"

7/�� �� :�/��.����.��!! ���/�����.�!���0"�567���"����>/��:�/�����/�/�>/�/����

A������������������ :4�3@���������!(�/�����/0!��@��/���������� ���>/��:�/��

@�����3������� 4�����������������@:�>/�/�������6"������+@6������ ���>/��:�/��

������ �����( 567�0/������!/������/���(������/!���/��.���!/0�!��������(���������������(�

>/��:�/���!������(�����������/0!�(�;<@����� !�

�&����� ���(4-�>/������ 4��.�����"������>���,�3&3--4����������.�������B#�&$$�

-!������/!��/.��"� 4��.�����"������-A� #$#$%#����������.�������B#�&$$�

-!������/������������.��/�����

4��.�����"������-A� #�& %#����������.�������B#�&$$�

��-��������

!�$ .�/����)01�0#

�

2

0

1

2

3

4

5

6

7

00:00 07:12 14:24 21:36 28:48

Illumination time [ min]

log1

0 TC

ID50

/ml LED device

M acot ronic

First investigations on a newly developed LED illumination device for the treatment of MB-plasmaU. Gravemann1, M. Behague2, S. Reichenberg3, H. Mohr1 , W. H. Walker3, T. Verpoort2, T.H. Mueller1

1Blood Center of the German Red Cross Chapters of NSTOB, Institute Springe, Germany2Maco Pharma S.A., Tourcoing, France 3Maco Pharma International, Langen, Germany

Treatment of fresh frozen plasma (FFP) with MB (methylene blue) and light is a procedure used for the inactivation of blood-borne viruses for more than 15 years. MB plasma is currently produced in several European countries using the MacoPharma Theraflex MB plasma system. High-intensity, long-lived light-emitting diodes (LEDs) are now available on the market, which might replace the sodium vapour lamps currently used in the Macotronic device (MacoPharma). The purpose of this study was the first evaluation of a newly developed LED illumination device (picture 1) with respect to virus inactivation capacity and plasma quality.

Treatment was done in the Theraflex MB plasma bag system (MacoPharma), containing leucocyte-depletion filter Plasmaflex, MB pill (85 µg MB), illumination bag, MB depletion filter Blueflex and plasma storage bag. For investigating plasma quality 3 plasma pools were each prepared from 3 different single donor units. Plasma was divided into three illumination bags and illuminated on the LED device. Samples were taken at different time points of illumination and factor VIII and fibrinogen (Clauss) were determined.

For investigating virus inactivation capacity Pseudorabies virus (PRV) was spiked into FFP (n = 4) resulting in a titer of approximately 106 tissue culture infectious doses/ml (TCID50/ml). The viral titer was determined from samples taken after 5, 10, 15 and 20 minutes of illumination on the LED device. Infectivity was determined by endpoint titration using a Vero cell CPE (cytopathic effect) assay. In a preliminary run the currently used Macotronic was compared with the newly developed LED device.

40th Annual Meeting of the German Society for Transfusion Medicine and Immunohaematology (DGTI) 2007

Virus inactivation was investigated using the PRV, an enveloped, double-stranded DNA virus, which is used as model virus for HBV. MB/light treatment using the LED-based illumination device resulted in an inactivation of > 4 log steps of PRV after 10 – 15 min of illumination (diagram 1). The device is at least as effective as the Macotronic device routinely used at present.

Plasma quality was only slightly affected by illumination. Factor VIII was decreased by 17% and fibrinogen (Clauss) by 14% during an illumination of 20 min (Diagram 2). Results are at least comparable to or might even be better than those for the Macotronic device.

Table 1: Inactivation kinetics of PRV (n = 4)

50 .0

60 .0

70 .0

80 .0

90 .0

100 .0

110 .0

0 5 10 15 20Illumination time [ min]

% o

f st

artin

g va

lue

FibrinogenFactor VIII

A new, compact illumination device based on long-lived LEDs was developed. The preliminary data suggest that this LED device is comparable to the Macotronic device with respect to virus inactivation capacity and preservation of plasma quality. Illumination time might even be shortened by using this high intensity illumination device.

Purpose

Methods

Results

detection limit

Diagram 1: Inactivation of PRV by LED device (n=4) and Macotronic (n=1) Diagram 2: Influence on coagulation factors (n = 3)

Conclusions

sampling time [min] 0 5 10 15 20

log10 TCID50/ ml (mean ± SD) 6.16 ± 0.23 4.64 ± 0.68 ≤ 2.41 ± 1.65 ≤ 1.75 ± 0.38 ≤ 1.54 ± 0.00

log10 reduction factor 1.52 ≥ 3.75 ≥ 4.41 ≥ 4.62

Picture1: Prototype

Picture 2: Pre-series device

3

Three years’ haemovigilance of methylene blue-treated fresh frozen plasma :no increase in transfusion reaction incidence

Deneys V.1,2, Lambermont M.2, Latinne D.1, Guerrieri C.1, Baele P.11 Tranfusion Committee, Cliniques Universitaires Saint Luc, 1200 Brussels, BELGIUM

2 Service Francophone du Sang, Belgian Red Cross, BELGIUM

BackgroundDue to stringent donor selection, laboratory testing and pathogen inactivation procedures, fresh frozen plasma (FFP) offers a high degree of viral safety. In Belgium, only non remunerated volunteers are recruited as plasma, platelets and blood donors. Pathogen inactivation is a proactive strategy designed to inactivate a pathogen before it enters the blood supply. Methylene blue-photoinactivation was choosen by the Belgian Red Cross two years ago (Theraflex MB plasma –MacoPharma)

MB-viroinactivation (Theraflex-MB plasma MacoPharma)

1. Plasma Connection(Sterile Docking device)

2. MicroFiltration : Removal of residual cells

3. Illumination : Oxydative Catalytic Reaction

MB

4. MB + MB MetabolitesRemoval : Azurs B,C,A

5. Freezing

Collection FreezingMaximum 18 hours

Results(a) : benign allergic reaction (n=4)

moderate allergic reaction (n=1)NHFR (n=1)

(b) : benign allergic reaction (n=5)moderate allergic reaction (n=3)

No TRALI episode (any previous pregnancy / transfusion = contra-indication for plasma donation)

0.14 %0.12 %Incidence of reaction

8 (b)6 (a)Number of adverse reactions after FFP Tf°

56605101Number of FFP units transfused

mid 2004 – 20052003 – 2004Period

BM-FFPSD-FFP

MethodsUntil mid 2004, only solvent-detergent FFP (SD-FFP) was used in our teaching hospital. BM-FFP began to be transfused to our patients thereafter. Since nearly 10 years, all transfusions of labile blood products are checked for the appearance or not of an adverse event.In this study, the incidence and the seriousness of adverse event after BM-FFP transfusion were compared with those observed with SD-FFP.

Discussion and conclusionNo significant increase in adverse event incidence was observed after BM-FFP transfusion (odds ratio 1.2) and the seriousness of these events was comparable. The clinical efficacy of both FFP was similar : both procedures have limited effects on coagulation factors (especially on fibrinogen and factor VIII for MB, on protein S and alpha 2-antiplasmin for SD).Finally, plasma pooling may have the undesirable effect of increasing the risk of transmitting viruses that are either resistant to the process or have escaped the virucidal process. The hallmarkof MB technology is that it allows the viral inactivation of single donor units of FFP, offering reassurance that no increased infectious risks are added due to pooling. This a crucial point in term of public health safety.

4

The

pro

ced

ure

usin

g m

ethy

lene

blu

e (M

B) t

o in

activ

ate

viru

ses

in t

hera

peu

tic p

las-

ma

is w

ell

esta

blis

hed

wor

ldw

ide.

It

incl

udes

mem

bra

ne f

iltra

tion

(Pla

smaf

lex,

0.

65 µ

m p

ore

size

), ad

diti

on o

f MB

(dry

pill

, 85

µg, r

esul

ting

in 1

µm

ole/

L at

266

ml),

illum

inat

ion

(ap

pro

x. 2

0 m

in,

590

nm),

and

filt

ratio

n of

MB

and

pho

top

rod

ucts

(Blu

efle

x).

Mor

e th

an

4 m

illio

n un

its

of

pla

sma

wer

e tr

ansf

used

w

ithou

t an

y un

usua

l ad

vers

e ev

ent

rep

orte

d.

Aim

of

this

stu

dy

was

to

pro

ve t

he t

oxic

olog

ical

saf

ety

of M

B,

its p

hoto

pro

duc

tsaz

ure

A,

B a

nd C

, an

d t

hat

of M

B-t

reat

ed p

lasm

a.

Ad

sorp

tion,

dis

trib

utio

n an

d e

xcre

tion

of 1

4C-l

abel

ed M

B f

ollo

win

g 24

h in

fusi

onw

ere

inve

stig

ated

at

a

dos

e le

vel

of

20m

g/kg

b

ody

wei

ght

(b.w

.) in

ra

ts.

Ob

serv

atio

n tim

e w

as 9

6 ho

urs.

Stu

die

s on

ter

atog

enic

effe

cts

wer

e d

one

by

intr

aven

ous

bol

us in

ject

ion

of M

B in

tora

ts a

nd B

eagl

e d

ogs.

MB

was

ad

min

iste

red

dai

ly t

o th

e d

ams

at 4

, 12

, 36

mg/

kgb

.w.

(rat

) and

2,

6, 1

8m

g/kg

b.w

. (r

abb

it).

In a

tole

ranc

e te

st 5

ml/k

g b

.w. o

f aut

olog

ous

light

-tre

ated

pla

sma

(1 o

r 10

µM M

B)

was

ad

min

iste

red

to

5 m

ale

Bea

gles

per

gro

up b

y in

trav

enou

s ad

min

istr

atio

n. A

fter

21d

3 d

ogs/

grou

p w

ere

trea

ted

aga

in a

nd s

acrif

iced

24

h la

ter.

Hem

atol

ogy,

clin

i-ca

l bio

chem

istr

y, a

nd e

lect

roca

rdio

gram

wer

e ex

amin

ed. A

com

ple

te h

isto

pat

holo

-gy

was

don

e.

MB

and

Azu

re A

/B/C

wer

e te

sted

in: b

acte

rial r

ever

se m

utat

ion

test

(Am

es t

est),

invi

tro

mam

mal

ian

cell

gene

mut

atio

n te

st,

in v

itro

mam

mal

ian

chro

mos

ome

aber

ra-

tion

test

with

hum

an ly

mp

hocy

tes,

in v

ivo

mic

ronu

cleu

s te

st w

ith r

at b

one

mar

row

and

per

iphe

ral b

lood

cel

ls (2

0m

g/kg

b.w

., 24

h in

fusi

on),

in v

ivo

unsc

hed

uled

DN

Asy

nthe

sis

test

in r

ats

(20

mg/

kg b

.w.,

bol

us in

fusi

on).

ME

TH

OD

S

OB

JEC

TIV

ES

RE

SU

LTS

Thre

shol

ds f

or n

o or

low

tox

ic p

rope

rties

whi

ch o

ccur

red

afte

r ad

min

istra

tion

of M

B in

pr

eclin

ical

stu

dies

var

ied

depe

ndin

g on

the

am

ount

of

MB

app

licab

le in

the

spe

cific

tes

tsy

stem

. The

y ar

e >

160

to 2

00,0

00 fo

ld h

ighe

r tha

n th

e es

timat

ed c

linic

al e

xpos

ure

of M

Baf

ter i

nfus

ion

of 6

uni

ts M

B-li

ght t

reat

ed p

lasm

a.

1. P

harm

acok

inet

ics

of 14

C-la

bele

d M

B a

fter

24h

infu

sion

wer

e de

term

ined

in

Tmax,

T1/2

αan

d T1

/2β

It in

dica

ted:

- bi

phas

ic e

limin

atio

n of

MB

with

an

initi

al h

alf-

life

of 3

min

and

a lo

nger

ter

min

al h

alf-

life

of 1

2.6

h (m

ale)

and

16.

0h

(fem

ale)

- le

ss th

an 1

% ra

dioa

ctiv

ity in

pla

sma

and

exam

ined

org

ans

- E

xcre

tion

of ra

dioa

ctiv

ity w

as a

lmos

t com

plet

e af

ter 9

6h

- no

acc

umul

atio

n or

sto

rage

of M

B

2. T

he n

o ob

serv

ed e

ffect

lev

el (

NO

EL)

for

the

fet

al o

rgan

ism

was

4m

g an

d 6

mg/

kgb.

w./d

ay in

rats

and

rabb

its.

3. C

last

ogen

ic e

ffect

s of

MB

and

Azu

re B

wer

e fo

und

in v

itro.

4. N

o ge

noto

xic

effe

cts

on b

one

mar

row

, pe

riphe

ral

bloo

d ce

lls a

nd h

epat

ocyt

es a

fter

appl

icat

ion

of 2

0m

g/kg

b.w

. MB

and

Azu

re B

.

5.

No

sign

s of

in

tole

ranc

e or

se

nsiti

zatio

n af

ter

infu

sion

of

1

µM

or

10µM

M

B

light

-tre

ated

pla

sma

befo

re re

mov

al o

f MB

and

pho

topr

oduc

ts w

ere

obse

rved

.

Saf

ety

Of M

ethy

lene

Blu

e Tr

eate

d P

lasm

aP

ohl

er P

1 , Le

usch

ner

J2 , G

rave

man

n U

1 , R

eich

enb

erg

S3 ,

Wal

ker

WH

3 , M

ohr

H1

1 DR

K N

STO

B,

Sp

ringe

, G

erm

any,

2 LP

T K

G,

Ham

bur

g, G

erm

any,

3 Mac

o P

harm

a In

tern

atio

nal G

mb

H,

Lang

en,

Ger

man

y

AA

BB

Ann

ual M

eeti

ng 2

007,

Ana

heim

, US

A

CO

NC

LUS

ION

S

1.W

illia

mso

n et

al.

2003

Tra

nsfu

sion

43:

1322

-132

92.

Poh

ler

et a

l. 20

04 T

rans

fus

Med

Hem

othe

r 31

(sup

pl 3

):1-8

4, P

S30

5

RE

FER

EN

CE

S

Figu

re 1

: G

loba

l sta

tus

of T

hera

flex

MB

-Pla

sma

use

Fig.

2 R

ecov

ery

of 14

C-la

belle

d M

ethy

lene

blu

e (M

B).

The

reco

very

was

exa

min

ed

in S

prag

ue-D

awle

y ra

ts f

ollo

win

g or

al

adm

inis

tratio

n (p

.o.)

and

24 h

i.v.

infu

sion

at a

dos

e le

vel o

f 20

mg

MB

/kg

body

wei

ght.

Urin

e, fa

eces

, org

ans,

exp

ired

air,

rinse

wat

eran

d in

fusi

on s

ite w

ere

anal

ysed

. The

radi

oact

ivity

reco

very

rate

in o

rgan

s an

d at

the

infu

sion

.

Fig.

3 M

ean

reco

very

of r

adio

activ

ity a

fter o

ral (

gava

ge) a

pplic

atio

n an

d 24

h i.

v. in

fusi

on o

f 20

mg

MB

/kg

body

wei

ght i

n ra

ts.

Fig.

4 S

afet

y m

argi

ns fo

r tox

icity

from

in v

ivo

stud

ies

with

em

thyl

ene

blue

.Cal

cula

tions

are

bas

ed o

n th

e tre

shol

d N

OEL

: no

obse

rved

effe

ct le

vel (

slig

ht m

etha

emog

lobi

nem

ia) (

12 w

eeks

tox

icity

) and

a n

orm

al c

linic

al

expo

sure

of

0,1

µg M

B/k

g bo

dy w

eigh

t (=

6 un

its

MB

pla

sma)

. 1 no

toxi

city

occ

ured

; the

refo

re th

e sa

fety

mar

gins

are

bas

ed o

n th

e hi

ghes

t dos

e te

sted

.

5

Background: During the last 15 years the method using methylene blue (MB) to inactivate viruses in plasma was constantly improved. Inventedby the Blood Center of the German Red Cross, chapters of NSTOB, Institute Springe, the initial procedure included: Freezing and thawing torelease intracellular viruses from leucocytes, addition of a proportional amount of a MB stem solution to a final concentration of 1 µM, and sub-sequent one-side illumination for one hour with fluorescent tubes.Aim: The aim was to improve the original method to facilitate the implementation in the blood bank.

Results

Conclusions: The MacoPharma Theraflex MB-Plasma represents an efficient, safe, and easy to use system which generates virus-safe plasma of high quality.

Results: With the current Theraflex MB-Plasma procedure provided by MacoPharma the procedure is markedly improved. The elimination of leucocytes is realized by membrane filtration, MB is added as an integrated dry pill, and residual MB and photoproducts are removed by a special Blueflex filter. The specially designed illumination device (Macotronic) ensures treatment under GMP conditions. Illumination dose and intensity are constantly monitored and temperature is controlled. The use of sodium low pressure lamps as improved light sources allowed thereduction of the illumination time to about 20 min.

The characteristic features of the system are:1. Virus inactivation of enveloped viruses shows a reduction rate of at least 5 log10 steps. (Figure 4)2. Plasma quality: Only fibrinogen and factor VIII are reduced by about 20-25%. (Figure 3)3. Clinical use: More than 4 million MB-treated plasmas were transfused with excellent tolerance and efficacy in

several countries all over the world. (Figure 6)4. MB and photoproducts are eliminated by more than 90% using the Blueflex filter. (Figure 7)5. Toxicology: Investigation on toxicology of MB and photoproducts showed a high safety margin for the concentration used. (Figure 5)

CHARACTERISTICS OF MACO PHARMATHERAFLEX MB-PLASMA

S. Reichenberg1, U. Gravemann2, P. Pohler2, W. H. Walker1

1Maco Pharma, Langen, Germany2Blood Center of the German Red Cross Chapters of NSTOB, Institute Springe

ISBT Congress, Madrid, June 2007

S ensitivity o f en ve lo ped viruses V irus Fam ily R eduction ra te

(log 10)

H IV -1 R etro 5 ,45 2

W N V F lavi 5 ,78 *2

BV D V F lavi 5 ,44 *2

H og cholera F lavi 5 ,92 *1

P R V H erpes 5 ,48 *2

H erpes S implex H erpes 5 ,50 *1

Bovine herpes H erpes 8 ,11 *1

S emliki Forest T oga 7 ,00 *1

S indbis T oga 9 ,73 1

Influenza O rthomyxo 5 ,1 1

H BV (D uck model) H epadna > 6 3

V esicu lar S tomatitis R habdo 4 ,89 *1

S ensitivity o f non -enve loped viruses V irus Fam iliy R eduction ra te

(log 10)

Adeno Adeno 4 1

C a lici C a lici 3 ,9 *1

S V 40 P apova 4 1

P arvo B 19 P arvo 5 1

1 M ohr et a l. Immunological Investigations 1995, 24(1&2); 73-85

2 tested by Analysis Biom edizinische Test G mbH 3 tested by P rof. C hristian TR E PO et a l., IN SE R M U nit 271, Lyon, F rance

* tested under production conditions R eduction below the lim it of detection

T e st U n i t R a n g eG lo b a l te stT h ro m b in tim e [s] 1 4 - 2 1 15 ,2 ± 0 ,3 1 7 ,4 ± 0 ,9 15 ,9 ± 1 ,1 1 6 ,2 ± 1 ,1 18 ,5 ± 0 ,5 20 ,2 ± 0 ,9C o a g u la tio n fa c to rsF ib rin o g e n (C la u ss) [m g /d l ] 20 0 - 4 50 26 2 ,3 ± 8 ,5 1 9 0 ,5 ± 12 ,6 19 4 ,0 ± 1 4 ,0 1 90 ,8 ± 1 1 ,3 25 4 ,5 ± 1 3 ,4 22 7 ,0 ± 1 1 ,7F a c to r I I [%] 7 0 - 1 30 10 4 ,8 ± 2 ,1 1 0 1 ,9 ± 2 ,6 96 ,6 ± 1 ,4 1 03 ,0 ± 5 ,3 10 9 ,3 ± 5 ,2 10 4 ,5 ± 2 ,9F a c to r V [%] 6 0 - 1 30 87 ,1 ± 6 ,0 1 0 5 ,6 ± 8 ,8 10 8 ,1 ± 3 ,8 1 05 ,0 ± 8 ,8 99 ,4 ± 6 ,4 10 7 ,6 ± 3 ,8F a c to r V I I I [%] 6 0 - 1 50 88 ,6 ± 17 ,9 7 2 ,5 ± 15 ,7 82 ,3 ± 1 7 ,6 7 2 ,9 ± 1 3 ,8 73 ,5 ± 1 4 ,8 81 ,5 ± 1 3 ,1F a c to r IX [%] 6 0 - 1 30 10 0 ,4 ± 5 ,8 9 2 ,8 ± 3 ,1 90 ,9 ± 6 ,8 9 7 ,8 ± 4 ,8 77 ,5 ± 5 ,3 99 ,6 ± 8 ,1F a c to r X I [%] 6 0 - 1 30 98 ,6 ± 5 ,4 7 8 ,4 ± 7 ,1 80 ,1 ± 4 ,5 7 9 ,3 ± 5 ,9 71 ,5 ± 4 ,3 87 ,5 ± 3 ,4v W F :R C o [%] 6 0 - 1 50 96 ,5 ± 5 ,3 1 0 0 ,5 ± 15 ,5 11 0 ,3 ± 2 0 ,7 1 12 ,8 ± 2 2 ,2 10 1 ,8 ± 1 5 ,8 11 0 ,8 ± 1 9 ,8In h ib i to rsfre e P ro te in S [%] 5 5 - 1 30 10 4 ,3 ± 6 ,4 1 0 3 ,5 ± 7 ,0 81 ,8 ± 7 ,2 9 8 ,8 ± 1 0 ,5 99 ,0 ± 5 ,9 98 ,8 ± 7 ,0P ro te in C [%] 7 0 - 1 40 97 ,8 ± 7 ,7 8 9 ,5 ± 5 ,9 85 ,0 ± 7 ,8 8 1 ,0 ± 1 8 ,5 11 4 ,0 ± 1 0 ,3 97 ,0 ± 6 ,5A T I I I [%] 8 0 - 1 20 91 ,3 ± 3 ,3 9 0 ,5 ± 3 ,3 95 ,5 ± 2 ,6 9 1 ,8 ± 3 ,8 11 0 ,8 ± 4 ,1 10 2 ,0 ± 6 ,8F ib rin o ly sisa 1-A n ti try p sin [m g /d l ] 9 0 - 2 00 98 ,5 ± 1 ,3 9 7 ,5 ± 2 ,5 98 ,3 ± 1 ,5 9 9 ,3 ± 2 ,1 10 6 ,5 ± 2 ,4 10 0 ,8 ± 3 ,4a 2-A n tip la sm in [%] 8 0 - 1 20 95 ,0 ± 2 ,8 9 4 ,0 ± 2 ,6 92 ,8 ± 4 ,4 8 4 ,8 ± 2 ,1 96 ,0 ± 4 ,2 10 0 ,5 ± 4 ,4A c tiv a tio nF a c to r X I Ia [m U /m l ] < 5 0 31 ,3 ± 4 ,1 3 3 ,3 ± 3 ,9 33 ,4 ± 6 ,2 3 5 ,1 ± 6 ,4 36 ,3 ± 6 ,8 36 ,3 ± 4 ,7C o m p le m e n tC H 1 0 0 [U /m l ] 39 2 - 1 01 9 68 9 ,6 ± 15 7 ,4 5 7 9 ,1 ± 31 ,7 94 9 ,3 ± 8 5 ,7 7 71 ,3 ± 1 61 ,5 79 8 ,8 ± 1 79 ,7 97 8 ,0 ± 7 4 ,1

1 8 m o n th 2 7 m o n thIn i t. v a lu e a fte r p re p . 3 m o n th 9 m o n th

Figure 1: bag system Figure 2: Illumination device Macotronic

Figure 4: Virus reduction capacityFigure 3: Plasma quality after treatment during storage for 27 month

Figure 5: Safety margin for toxicity of Methylene Blue Figure 6: Global status of Theraflex MB-Plasma use Figure 7: Methylene Blue reduction by Blueflex filtration

Virus reduction capacity

Plasma quality

Safety margins for toxicity from in vivo studieswith MB and MB-treated plasma Methylene blue reduction

Safety margins for toxicity from in vivo studies with methylene blue. Calculations arebased on the thresholds NOEL: no observed effect level (teratology) and LOEL: lowestobserved effect level (slight methaemoglobinemia) (12 weeks toxicity) and a normal clinical exposure of 0.1 µg MB/kg body weight (= 6 units of MB plasma). 1No toxicity occurred, therefore the safety margins are based on the highest dose tested.

Pohler et al. 2004 Transfus Med Hemother 31 (suppl 3):1-84, PS305

Reichenberg et al. 2006 Vox Sang 91 (suppl 3): P386

Conclusions6

THE EFFECT OF METHYLENE BLUE PATHOGEN REDUCTION SYSTEM ON Fc VIIIc IN PLASMA DERIVED FROM WHOLE BLOOD DURING STORAGE

Baeten M1, Rapaille A2, Donnay D2, Van der Beek M1, Sondag D2, Vandekerckhove P 1.1 Dienst voor het Bloed, Rode Kruis-Vlaanderen, Belgium.

2 Service du Sang, Croix Rouge de Belgique, Belgium.

Regarding the Fc VIIIc activity of plasma, Methylene blue pathogen reduction has to be completed within a limited time interval after whole blooddonation. The processing / separation of whole blood can be performed at any time between donation and the photo treatment of plasma. Following the attainment of these results, methylene blue pathogen reduction of plasma has been implemented in both centres.

C O N C L U S I O N

The virucidal properties of methylene blue have been documented since 1930. In 1991, the Springe Institute developed a photodynamic method toinactivate pathogens, particularly viruses, in human plasma using methylene blue in combination with visible light. MacoPharma has improved thismethod and developed the THERAFLEX MB-Plasma system consisting of the Macotronic illumination machine, together with an appropriate disposable set for pathogen reduction and removal of residual methylene blue to a level less than 4µg/unit. This method is known to be effective on viruses as well as other documented pathogens, although reducing slightly the activity of clottingfactors such as factor VIII (Fc VIIIc) and Fibrinogen. According to Belgian legislation, plasma should be frozen within 18 hours after blood collection.In the case of pathogen reduced plasma, a level of at least 0.5 IU / ml for Fc VIIIc should be attained.The aim of this study was to investigate the effect on Fc VIIIc recovery of various time delays, between donation and the photodynamic treatment of plasma derived from whole blood.

B A C K G R O U N D

In centre 1, 143 units of whole blood were selectedfrom volunteer male A+ donors. These units weredivided into 3 groups. Plasma from group A wasseparated at 4 hours and treated at 5.5 hours,group B plasma at 4 and 16.5 hours, and group Cplasma at 15 and 16.5 hours, respectively.

In centre 2, 120 units of whole blood were selectedfrom volunteer male or female donors of any bloodgroup. These units were also divided into 3 groups.Plasma from group D was separated at 3.5 hoursand treated at 8 hours, group E plasma at 3.5 and11 hours, and group F plasma at 12.5 and 16.5, respectively.

Pathogen reduction was performed using the THERAFLEX MB-Plasma system (disposable with

leucodepletion filter and methylene blue removalfilter, Macotronic illumination machine). Samplesfor Fc VIIIc activity assay were taken immediatelyafter separation and after treatment. Fc VIIIc measurement was done using a one-stage aPTTclotting assay with Fc VIIIc deficient plasma. Resultsof Fc VIIIc recovery are expressed in percentage of activity.

Prior to separation and photo treatment, the wholeblood and plasma were stored on eutectic plates tokeep the products at a temperature of 20 °C.Results were analysed using repeated measuresAnova for general comparison, student t-test forgroup comparison and paired student t-test whenappropriate.

M A T E R I A L S A N D M E T H O D S

- The difference between the two groups was statistically significant (p<0.001).- A significant decline in Fc VIIIc activity was measured in all groups following the phototreatment process (p<0.001).- No significant difference between group B and groups C and F after photo treatment wasobserved (p=0.87, p=0.10); this suggests no significant difference in the loss of Fc VIIIcactivity between the time of separation of whole blood into plasma and photo treatment 16.5 h.- No significant difference between Fc VIIIc activity loss in group A and C (p=0.66) and ingroup D and F (p=0.53) suggesting that the time interval between blood collection andseparation does not influence the loss of Fc VIIIc activity post photo treatment.

Fc VIIIc pre- and post-photo treatment mean results are presented in table 1.

R E S U L T S

Group A 4 5.5 85 67

Group B 4 16.5 92 56

Group C 15 16.5 71 56

Group D 3.5 8 83 73

Group E 3.5 11 81 66

Group F 12.5 16.5 75 64

Time beforeseparation (h)

Time beforetreatment (h)

Pre treatment 1

Fc VIIIc (%)Post treatment 2

Fc VIIIc (%)Fc VIII activity loss after storage at room temperature

and phototreatment

40

50

60

70

80

90

100

0 5 10 15 20

hours after collection

Fc

VIII

% a

ctiv

ity

Group AGroup BGroup CGroup D

Group EGroup F

60303 0484847 6 03030484847N =

GROUP

fedcba

FA

CT

VIII

300

200

100

0

-100

TIME

separ

inact

A B C D E F

SAMPLESBefore separation

After treatment

Table 1 : 1 Sample taken after separation 2 Sample taken after treatment

XE7R

A01A

07

/04

7

Quality Control Evaluation of Methylene Blue Light

Treated Plasma. L Larreal, A Cerveroz, M Calabuigl, A Blanquerl, P Solvesl, V Mirabet], R J Roigl.

~Centro De Transfusion De La Comunidad Valenciana, Valencia, Spain; 2H General Universitario, Valencia, Spain.

INTRODUCTION

Methylene blue light treated plasma has been used in Spain from 1998. Since then slight variations of the technique have been implemented such as plasma leukoreduction instead of freezing and thawing. Studies on plasma quality have been published analysing initial methods. We have tested several plasma units to evaluate plasma quality with the current technology.

MATERIAL AND METHODS

Briefly speaking, 450150 mL whole blood units were collected in automatic scales using top & bottom blood bugs. After collecting, units were cooled with 1.4- butanodiol plates and Iater stored ut 2212"~ Then, whole blood was centrifuged ut high speed to obtain a concentrate of red blood cells and plasma, while maintaining the buffy coat in the initial blood bug. For plasma inactivation the Springe modified method was used (Theraflex-MB- Plasma system: Macopharma@). Plasma was joined to the MB system by means of sterile docking and, simultaneously, gravity filtered. In batches of four plasma wu

r - illuminated for 20 min. Units for storage were frozen after inactivating before 24 hour postdonation. For the study we inactivated 30 plasma units (10 A, 10 0,5 B and x-~ * 5 AB). We took several samples: before inactivation (samplel), just after inactivation ._ 1

(sample2), after 1 month of storage ut -30% (sampleS), after 3 months of storage ut -30% (sample4) and after 6 months of storage ut -30% (sample5). After each moment samples were stored ut -80% till the tests were performed. In every sample we performed the following tests: PT, APTT, FV, FVlll and fibrinogen.

The process of the 5 plasma samples

INACTIVATION

I I I TlME

Sample 1 Sample 2 Sample 3 Sample 4 Sample 5

RESULTS

As published before most affected parameters by the inactivation procedure were fibrinogen and FVlll (18 and 16% respectively decrease from sample 1 to 2). FV was scarcely affected (a 3% decrease from sample 1 to 2). PT an APTi were prolonged only in 2.74 and 5.26% respectively from sample 1 to 2. Results may be seen with more detail in the attached table.

Results of the tests (PT, APTT, Fib, FV and FVIII) for the 5 plasma sample

A m : Activated Partial Thromboplastin Time test FVIII: Factor Vlll

CONCLUSIONS The Methylene blue inactivation methodology is very easy to use and the plasma factors after inactivation are preserved. However, during storage there is a certain loss of coagulation factors. If the reason for this is related to the treatment or the storage conditions remains to be evaluated.

Fib: Fibrinogen

8

THERAFLEX MB-PLASMA PROCEDURE: PLASMA QUALITY AFTER 15 MONTH STORAGE

S. Reichenberg1, W. Walker1, A. Hoburg2, N. Müller2

1MacoPharma International GmbH, Langen2Institute for Transfusion Medicine, University Hospital Essen

ISBT Congress, Cape Town, September 2006

Background: Although in the last decades thanks to the implementation of several methods like donor selection and testing procedures the risk of virus transmission from plasma has decreased, infection of patients still exists. Additionally new viruses like West Nile Virus enter the transfusion chain. Therefore, thetreatment of therapeutic plasma with methylene blue (MB) is a technique used in several European countries for virus inactivation. MacoPharma has developed the proprietary photodynamic Theraflex MB-Plasma bag system including a MB pill, an illumination system (Maocotronic) with visible light, and a finalMB filtration step with the Blueflex filter (Williamson et al. Transfusion 2003;43:1322-1329).

Aims: Aim of this study was to show the reduction of MB and photoproducts due to the Blueflex filter and to prove the reproducibility of the filtration efficiency.Additionally the quality of the plasma after 15 month storage was checked.

Introduction

Materials & Methods

Results

Conclusions

18 plasmas were treated at three different days. At different steps of the Theraflex MB-Plasma procedure the MB and photoproduct content was measured by HPLC, whichwas described previously (Verpoort et al. 2003; ISBT Istanbul P246). Measurement was done after dissolution of the MB pill, after illumination, and after filtration (Fig. 2).

At each day plasma was pooled and divided into several storage bags (storage temperature <-30 °C). At different time points a palette of plasma factors was measured.1. global tests (Quick, INR. aPTT, thrombin time)2. coagulation factors (Fibrinogen, F II, F V, F VII, F VIII:c, F IX, F X, F XI, F XII, FXIII, vWF Ristocetin Co-Factor)3. Inhibitors (AT III, Protein C, Protein S) 4. Fibrinolysis (Plasmin inhibitor, alpha-1-Antitrypsin, Plasminogen)5. Complement (CH50, C3a)6. Activation (TAT, F XIIa, D-Dimer)

Illumination of MB-containing plasma with visible light using the Macotronicillumination device resulted in the generation of photoproducts as describedpreviously. Mean reduction of the total phenothiazine content was 94.5%.Every single filtration yielded in a filtration efficiency of minimum 91%. Themean reduction capacity for MB was above 99.9%.

There was no significant change in the plasma factor content after treatmentduring the whole 15 month storage period. Slight variations are within theerror of measurement. The only difference between the plasma parametersresulted from the treatment itself. Here, an increase in the INR and aPTT(14.3 %; 18.2 %), decrease of fibrinogen (-19.3 %), factor V (-25.3 %),factor VIII (-21.8 %), factor IX (-25.6 %), factor X (-23.4 %), and factor XI(-16.8 %) was observed. Despite this variations the values were within theranges found in non-treated plasma.

The filtration of plasma with the Blueflex filter is a reliable method to reduce the amount of MB and photoproducts substantially.The plasma quality is not changed during the observed storage period ad remains within the physiological variation of non-treated plasma.

Test Limits Unit 0 month 6 month 9 month 15 month

1. Gobal tests

Quick 80 -130 % 98 ± 2 105 ± 3 94 ± 2 93 ± 3INR 1.0 ± 0.0 1.0 ± 0.1 1.1 ± 0.1 1.1 ± 0.1

aPTT 30 - 60 sec 34 ± 1 35 ± 1 34 ± 1 35 ± 1Thrombin time 14 - 21 sec 22.7 ± 0.7 23 ± 1.0 23 ± 1.0 22 ± 1.0

2. Coagulation factors

Fibrinogen 1.8 - 3.5 g / l 2.4 ± 0.1 2.3 ± 0.2 2.4 ± 0.2 2.4 ± 0.1F II 70 - 130 % 99 ± 4 104 ± 4 106 ± 7 105 ± 12F V 65 - 150 % 101 ± 3 99 ± 6 103 ± 9 101 ± 10

F VII 70 - 130 % 103 ± 8 119 ± 9 108 ± 3 102 ± 4F VIII:c 050 - 2.00 I.U. / ml 0.81 ± 0.15 0.83 ± 0.12 0.92 ± 0.16 0.87 ± 0.11

F IX 0.70 - 1.30 I.U. / ml 1.00 ± 0.05 0.92 ± 0.04 0.98 ± 0.04 0.94 ± 0.04F X 70 - 130 % 106 ± 6 108 ± 5 111 ± 7 98 ± 3

F XI 50- 130 % 82 ± 2 92 ± 3 83 ± 3 83 ± 4F XII 70 - 130 % 97 ± 10 101 ± 11 99 ± 10 98 ± 11F XIII 70 - 130 % 81 ± 11 77 ± 7 75 ± 4 85 ± 9

vWF (Ristoc. Co-F.) 50 - 140 % 85 ± 8 81 ± 19 94 ± 13 95 ± 7

3. Inhibitors

AT III 0.80 - 1.30 I.U. / ml 1.12 ± 0.08 1.02 ± 0.08 1.06 ± 0.07 0.99 ± 0.06Protein C 70 - 150 % 110 ± 7 109 ± 6 116 ± 6 110 ± 3Protein S 70 - 140 % 75 ± 7 70 ± 2 76 ± 2 81 ± 2

4. Fibrinolysis

Plasmin inhibitor 80 - 120 % 103 ± 4 101 ± 3 103 ± 6 105 ± 10alpha-1-Antirypsin 0.70 - 1.50 I.U. / ml 1.14 ± 0.03 1.12 ± 0.07 1.24 ± 0.09 1.20 ± 0.08

Plasminogen 75 - 140 % 102 ± 7 105 ± 8 12± ± 11 104 ± 10

5. Complement

CH50 70 - 100 % 116 ± 14 110 ± 11 108 ± 5 114 ± 8C3a 123-2228 ng / ml 1029 ± 323 1106 ± 596 898 ± 332 921 ± 217

6. Activation

TAT 1 - 4.1 µg / l 2.1 ± 0.2 2.4 ± 0.5 2.0 ± 0.0 2.2 ± 0.7F Xlla < 3.0 ng / ml 1.0 ± 0.2 0.9 ± 0.1 1.3 ± 0.6 1.0 ± 0.0

D-Dimer 64 - 246 µg / l 241 ± 120 239 ± 131 199 ± 97 225 ± 122

Fig. 1: Storage stability of Theraflex MB-Plasma

Fig. 2: Methylene blue and photoproduct reduction due to Bluflex filtration

9

QUALITY OF THERAFLEX® MB-PLASMA DURING STORAGE AND TREATMENT

N. Müller1, S. Reichenberg2

1Institute for Transfusion Medicine, University Hospital Essen2MacoPharma International GmbH, Langen

ISBT Congress, Athens, July 2005

Background: Although in the last decades thanks to the implementation of several methods like donor selection and testing procedures the risk of virustransmission from plasma has decreased, infection of patients still exists. Additionally new viruses like West Nile Virus enter the transfusion chain [1].Therefore, the photodynamic treatment of therapeutic plasma with methylene blue (MB) is a technique used in several European countries for pathogeninactivation [2]. MacoPharma has developed the proprietary Theraflex® MB-Plasma bag system including a MB pill and a final MB filtration step.

Aim: Aim of the study is to show the quality of the MB plasma during the preparation procedure and during storage using the Theraflex® system (see Figure 1).

Introduction

Materials & Methods

Results

Conclusions

Preparation ProcessFor the preparation process every single step was evaluated using 18 singledonor plasma units. For the evaluation of the plasma factors 5 ml were drawnat different stages (see Figure 2). The samples were pooled after drawing andmeasured for the specified factors. Six samples of each stage were pooled atthree days. A whole panel of plasma factors was measured for the resultingthree pools (see Figure 3).StabilityStability data were generated using three plasma pools. Six plasmas were pooled and afterwards divided into six aliquots. Each was treated as single unitand then each was divided into six storage samples. The same plasma factorsas for the manufacturing process were evaluated.MB and photoproduct content was below the detection limit as previously described [3].

All investigated plasma factors remained stable during the investigated storage time. A moderate reduction for some coagulation factorsduring the preparation was found in the illumination step but not in the other preparation stages. This was mainly fibrinogen (17,5 %), factor VIII (22,2 %), and factor X (13,4 %). Despite this reduction the values were within the ranges found in non-treated plasma.

In-process control Storage stability

Plasma treated with the Theraflex procedure showed slight reduction during treatment and no reduction during storage. All plasmafactors remained within the threshold values. The treatment of therapeutic plasma with MB is a valid technique of pathogen inactivation.

[1] West Nile virus in plasma is highly sensitive to methylene blue-light treatment Mohr et al, Transfusion 2004;44:886-890[2] Methylene blue-treated fresh-frozen plasma: what is its contribution to blood safety? LM Williamson et al, Transfusion 2003;43:1322-1329[3] Filtration of Methylene Blue and Photoproducts after Photodynamic Treatment of Plasma using BLUEFLEX T Verpoort et al, 2003 VIII European ISBT Congress P245

Figure 3: Percentage of deviation from the source plasma for different plasma factors during storage and treatment A: before treatment; B: after PLAS4 filtration; D: after MB addition; D: after illumination; E: after Blueflex filtration

Figure 1: Theraflex® MB-Plasma bag system

Figure 2: Sampling scheme

Global tests

Coagulation factors

Coagulation inhibitors

Complement activation

10

Blood Center of the German Red Cross Chapters of NSTOB

PREPARATION OF METHYLENE BLUE –TREATEDPLASMA UNDER WORST-CASE CONDITIONS- INFLUENCE ON QUALITY AND STABILITY-Gravemann U1, Pohler P1, Reichenberg S2, Budde U3 , Walker W2, Mohr H1, Müller TH1

1 Blood Center of the German Red Cross, Chapters of NSTOB, Institute Springe, Germany 2 MacoPharma International, Langen, Germany3 Coagulation Laboratory, Laboratory Prof Arndt&Partners, Hamburg, Germany

INTRODUCTION

Treatment with methylene blue (MB) and light is a well-known procedure for the inactivation of blood-borne viruses in Fresh Frozen

Plasma (FFP). The purpose of this study was to assess the quality and stability of MB/light-treated plasma (MB plasma) processed by

the MacoPharma Theraflex MB-Plasma® system. Preparation was done under worst case conditions for routine processing to evaluate

the worst plasma quality to be expected during production.

CONCLUSIONS

Even under worst-case conditions, photodynamic treatment of FFP using the Theraflex MB-Plasma® system only moderately affects

the activities of coagulation factors. The Blueflex-filter depletes MB and its photoproducts by over 90% after photodynamic treatment.

Storage of MB plasma for up to 9 months had no effect on coagulation factors.

METHODS

12 single donor units of MB/light treated plasma were prepared using the MacoPharma Theraflex MB-Plasma® system. Preparation

included leukocyte depletion (Plasmaflex-filter), addition of methylene blue (MB-pill) prior to illumination and depletion of MB and

photoproducts (Blueflex-filter) after treatment. Samples were taken before treatment and from the final product. For the assessment of

stability, plasma from four different plasma pools was photodynamically treated and stored for up to 9 months. Treatment was done

under worst-case conditions for the preservation of product quality: maximum MB concentration during illumination (1.15 µmol/l),

maximum storage time of whole blood before separation (4°C, 17 h), maximum storage time of MB plasma before freezing (1 h).

Parameter Before treatment After treatment Percentage of loss (-)

or increase (+)

Thrombin time [s] 15.8 +/- 0.7 19.1 +/- 1.5 +20.6 §

Fibrinogen (Clauss) [mg/dl] 279.9 +/- 53.6 222.5 +/- 41.8 -20.3 §

Factor V [%] 120.7 +/- 30 101.8 +/- 29.8 -16.4 §

Factor VIII [%] 115.2 +/- 23.9 89.5 +/- 21.3 -22.2 §

Factor XI [%] 94.6 +/- 20.1 82.1 +/- 18.7 -13.3 §

Antithrombin III [%] 87.6 +/- 7.4 87.3 +/- 6.0 -0.3 ns

Protein C [%] 105.3 +/- 21.3 94.5 +/- 17.4 -9.8 §

Protein S, free [%] 94.2 +/- 11.2 94 +/- 12 -0.2 ns

vWF:RCo [%] 98.8 +/- 23.5 98.8 +/- 23.5 -5.0 ns

vWF CP [%] 52.8 +/- 13.9 50.2 +/- 9.7 - 1.8 ns

Plasmin inhibitor [%] 94.2 +/- 9.1 93 +/- 8.7 -1.2 ns

α1-Antitrypsin [mg/dl] 95.8 +/- 19.3 95.3 +/- 19 -0.5 ns

F 1+2 [nmol/l] 1.02 +/- 0.25 1.03 +/- 0.27 +1.0 ns

D-Dimers [mg/l] 0.20 +/- 0.04 0.18 +/- 0.04 -9.2 ns

CH 100 [U/ml] 618 +/- 115 633 +/- 169 +4.1 ns

DGTI Congress, Erfurt, 2005

Methylene Blue µmol/l Depletion (C / A) A (before illumination) 0.97 +/- 0.06 B (after illumination) 0.65 +/- 0.15 C (after Blueflex filtration) 0.01 +/- 0.01 98.8 % Azure A µmol/l A 0.01 +/- 0.01 B 0.06 +/- 0.03 C 0.01 +/- 0.01 Azure B µmol/l A 0.11 +/- 0.01 B 0.24 +/- 0.04 C 0.01 +/- 0.01 Azure C µmol/l A 0.00 +/- 0.00 B 0.03 +/- 0.03 C 0.01 +/- 0.02 Phenothiazine total µmol/l A 1.09 +/- 0.07 B 0.99 +/- 0.08 C 0.04 +/- 0.03 96.4 %

RESULTS

Thrombin time, fibrinogen (Clauss), factors V, VIII, XI and protein C were significantly altered by MB/light treatment, while anti-

thrombin III (AT III), vWF:RCo, vWF cleaving protease (vWF CP), plasmin inhibitor and α1-antitrypsin remained unchanged (Fig. 1).

There was no activation of the coagulation markers (F 1+2, D-dimers) attributed to the virus inactivation procedure including the

filtration steps for leukocyte depletion and MB and photoproduct depletion. The influence of each manufacturing step on the activity of

coagulation factors was investigated using three plasma pools. Most of the activity was lost during illumination (Fig. 2). After

illumination MB and its photoproducts (azure A, azure B, azure C) were depleted by Blueflex filtration (Fig 3) to a final concentration of

<0.1 µmol/l (MB + sum of photoproducts). Stability of MB-Plasma was tested during storage at -30°C for up to 9 months (Fig 4). Stability

testing will be continued for a total of 27 months.

0

20

40

60

80

100

120

after 1 week after 3 months after 9 months

[%]

Factor V Factor VIII Factor XI AT III

n = 12, § significant , P < 0.01 (Wilcoxon Rank Sum Test) ns not statistically significant

Fig.3 Depletion efficacy of the Blueflex filter(data from 12 single donor units)

Fig.1 Influence of the MB/light treatment on plasma quality(data from 12 single donor units, treatment under worst-caseconditions)

Fig.4 Stability of MB/light-treated plasma(data from 4 pools, , treatment under worst-case conditions)

Fig.2 Influence of the individual manufacturing steps on plasma quality(data from three pools)

50

60

70

80

90

100

110

120

130

140

Astartingmaterial

Bafter

Plasmaflex-filtration

Cafter

dissolution ofthe MB pill

Dafter

illumination

Eafter

Blueflex-filtration

Ffinal productafter thawing

% o

f st

artin

g va

lue

Thrombin time Fibrinogen (Clauss) Factor V Factor VIII

11

XI2IA03E 26/09/2005

For each stage in the

preparation

Samp.A Samp.B Samp.C Samp.D Samp.A Samp.Ab Samp.BParameter Limits (n = 40) (n = 40) (n = 10) (n = 10) (n = 10) (n = 10) (n = 10)

Prothrombin rate (%) 70 - 130 80.8 71.8 87.7 76.6 73.5 68.2 65.7 INR 1 1.21 1.33 1.11 1.24 1.3 1.39 1.43 Activated partial thromboplastin time test (ratio)

1.12 1.26 1.05 1.18 1.13 1.22 1.27

Fibrinogen 2 - 4 3.11 2.29 2.94 2.04 3.03 2.28 2.36 Factor II (%) 70 - 120 102 98 98 95.1 95.1 86.7 92.3 Factor V (%) 70 - 120 102.2 94.1 89.5 83.2 105.7 98.1 96.4 Factor VII (%) 70 - 130 113.1 101.9 106.1 92.1 107.6 100.6 104.2 Factor VIII (%) 60 - 150 100.6 73.5 101.1 76.1 114 89.5 86.1 Factor IX (%) 60 - 150 96.5 78.8 98.6 86 99.2 78.9 85.7 Factor X (%) 70 - 120 105.9 95.3 106.9 92.6 97.3 91.7 92.3 Factor XI (%) 60 - 140 90.9 75.3 86.7 64.4 85.5 70.8 65.7 Factor XII (%) 60 - 140 103.7 92.6 110.5 99.9 98.4 96 93.9 Antithrombin III (%) 80 - 120 104.8 89.9 104.4 101.9 105.8 103.7 93.9 Protein C (%) 70 - 140 110.9 105.3 108.4 103 120.7 112.9 112.6 Protein S (%) 70 - 140 82.6 78.6 81.6 71.2 83.8 77.6 81.4 V Willebrand Factor CoF ristocetin (%) 60 - 150 97.6 92.9 87.4 84 143.6 137.4 138.6

Von Willebrand Factor Ag (%) 60 - 150 134.6 118.2 130.3 128 143.6 121.8 122.2

Plasminogen (%) 80 - 120 102.7 100.9 100.8 96.9 103.6 101.6 101.6 α2-antiplasmin (%) 80 - 120 111.9 107.7 106.2 104.2 112.5 106.5 107.7 C3a (mg/l) 100 - 400 134.8 134.1 146.8 143.1 124.2 125.5 124.0 C5a (µg/l) 0.9 - 15.4 10.9 14.1 15.1 23.9 7.5 8.3 10.9 Factor XIIa (ng/ml) < 3.0 2.39 2.22 2.87 2.63 2.57 1.79 2.49 F1 + 2 prothrombin (nmol/l) 0.4 - 1.1 0.99 1.20 0.88 0.87 0.79 0.72 0.78

Platelet factor 4 (UI/ml) 56 - 805 317.2 301.1 413.9 375.4 186.9 197.2 208 A : Plasma before contact with methylene blue Ab : Plasma after visible light (plasmas 31 to 40) B : MB-removed Plasma before freezing C : MB Plasma after 6 months at - 30° C D : MB-removed Plasma after 6 months at - 30° C

THERAFLEX � MB PLASMA

Coagulation factors and activation parameters

12

XI2IA04G 11/09/2007

S+

N

NCH3

CH3

NCH3

CH3

S+

N

NCH3

CH3

NCH3

CH3

THERAFLEX-MB Plasma

Processing principle

- Intercalation of MB into nucleic acids - Excitation of MB by visible light - Oxidation of Guanosine - Degradation of nucleic acids

The combined action of Methylene Blue and light is a photodynamic process which blocks transcription and replication of viral RNA and DNA.

Illumination of plasma + Methylene Blue (590 nm, 180J/cm2)

Methylene Blue molecule

MacoPharma Methylene Blue Pill (85µg / unit of plasma)

Photo-inactivation procedure of THERAFLEX-MB PLASMA

200-315ml of plasma

(Aphaeresis or Whole Blood)

Filtration of plasma and

dissolution of the MB Pill

Illumination of plasma+MB with the Macotronic

V4

MB removal by filtration with

Blueflex

Plasma freezing

13

XI2IA04G 11/09/2007

Whole Blood Plasma with MB treatment

Whole Blood Plasma with MB treatment and MB removal

Aphaeresis Plasma : distribution in 2 units

Aphaeresis Plasma : distribution in 3 units

THERAFLEX-MB PlasmaExamples of systems and treatments

SDV system : plasma filtration with Plasmaflex > MB Pill dissolution > illumination > MB removal with Blueflex

Aphaeresis plasma > 690 mL*

SPV system : filtration of the aphaeresis unit with Plasmaflex MB Pill dissolution and distribution in 2 units > illumination > MB removal with Blueflex

Aphaeresis plasma Min 485 mL – max 650mL*

BSV + ZDV system : filtration of the aphaeresis unit with Plasmaflex and distribution in 3 units with the BSV system. Connection of each unit to a ZDV system : MB Pill dissolution > illumination > MB removal with Blueflex

BSV system : plasma filtration with Plasmaflex > MB Pill dissolution > illumination

* Volume ranges for plasmas to be treated with THERAFLEX-MB Plasma, based on process requirements.

Whole blood plasma Min 250mL - max 330 mL*

Aphaeresis plasma Min 500 mL – max 665mL*

SDV system : plasma filtration with Plasmaflex > MB Pill dissolution > illumination > MB removal with Blueflex

14

1322 TRANSFUSION

Volume 43, September 2003

Blackwell Science, LtdOxford, UKTRFTransfusion0041-11322003 American Association of Blood BanksSeptember 2003439Editorial

METHYLENE BLUE-TREATED FRESH-FROZEN PLASMAWILLIAMSON ET AL.

ABBREVIATIONS:

APTT

=

activated partial thromboplastin time;

FFP

=

fresh-frozen plasma; MB

=

methylene blue; MBFFP

=

methylene blue-treated fresh-frozen plasma; PT

=

prothrombin

time; TTP

=

thrombotic thrombocytopenic purpura.

From the University of Cambridge and National Blood Service,

Cambridge; the National Blood Service, Brentwood; and the

Scottish National Blood Service, Edinburgh, United Kingdom.

Address reprint requests to:

Lorna M. Williamson, University

of Cambridge and National Blood Service, Long Road, Cambridge

CB2 2PT, United Kingdom; e-mail: lorna.williamson@nbs.nhs.uk.

Received for publication February 27, 2003; accepted March

11, 2003.

TRANSFUSION

2003;43:1322-1329.

R E V I E W

Methylene blue-treated fresh-frozen plasma: what is its contribution to blood safety?

Lorna M. Williamson, Rebecca Cardigan, and Chris V. Prowse

ith current donor-selection criteria andvirus genome testing, fresh-frozen plasma(FFP) in the developed world is probablysafer than it ever has been. In the UK,

where FFP is not manufactured from first-time or lapseddonors, it has been estimated that the residual virus risksfrom a single unit of FFP are 1 in 10 million for HIV, 1 in50 million for HCV, and 1 in 1.2 million for HBV (Eglin R,written communication, January 2003). Against theselevels of risk, it has been questioned whether pathogenreduction of FFP is a necessary strategy and/or the bestuse of healthcare resources.

1

However, the appearance ofWest Nile virus in blood components in the US in 2002,with fatal transmissions in immunocompromised recipi-ents,

2

reminds us that sometimes viruses move ahead ofour ability to test for them. Also, background viral inci-dence in a population can change, as is currently observedin Scotland, with HIV levels showing an increase to threeper million population (Soldan K, written communica-tion, February 2003). It is now over 10 years since a pho-todynamic system using methylene blue (MB) and visiblelight was developed in Springe, Germany, for virucidaltreatment of FFP. The method has been used at varioustimes since then in Germany, Denmark, Portugal, Spain,and the UK, so it is timely to review its potential contribu-tion to overall FFP safety.

MB is a phenothiazine compound (Fig. 1), which wasfirst used clinically by Paul Ehrlich in the 1890s and hasbeen used to kill viruses since work at the Walter Reed

W

Hospital in the 1950s.

3

When activated by visible light, MBgenerates reactive oxygen species, mainly singlet oxygen,through a Type II photodynamic reaction, and it is thesethat are responsible for its pathogen inactivating proper-ties.

3-5

The original system developed in Springe, Germany,used an initial freeze-thaw step to disrupt intact WBCs,then added an amount of MB solution calibrated to theweight of the plasma pack, to achieve precisely the sameMB concentration in every pack. Later systems (Baxterand Macopharma) developed for small-scale use in bloodcenters involve sterile connection of the plasma pack(before or after freezing) to a pack with a WBC-reductionfilter upstream of a liquid pouch or a dry pellet containing85 to 95

m

g of MB (Fig. 2). To achieve the desired final MBconcentration of 1

m

M

, the input plasma volume has to bewithin a 200-to-300-mL range, so 600-mL apheresis unitsrequire splitting. In both the Springe and commercial sys-tems, the MBFFP packs are then exposed to visible wave-lengths of light to activate the MB. Because it is notpossible to use the equivalent of radiation-sensitive labelsto confirm illumination, the light-exposure system mustbe designed to ensure good manufacturing practice(GMP)-compliant control of both light intensity and dura-tion. Radio-frequency chips for this purpose are in devel-opment. During illumination, MB is converted to itsbleached leuko- form and to demethylated components(azure A, B, and C, and thionine; Fig. 1). A recent featurehas been the development of commercial filters for post-treatment MB removal, which reduce the residual MBconcentration to 0.1 to 0.3

m

M

. The plasma is then readyfor freezing or refreezing.

One of the attractions of the technique is that it isapplied to single units of FFP, without the need for pool-ing. Commercial systems are available that can be set upin standard blood center GMP conditions, without theneed to install specialized plant, and it is this model thatis in operation in the UK. Plasma is frozen locally, sent toone of three central MB-treatment points, then returnedfor distribution to hospitals.

PATHOGEN-REDUCTION SPECTRUM

The ability of MB to inactivate viruses is dependent on itsbinding to nucleic acid, being greater for double stranded

15

METHYLENE BLUE-TREATED FRESH-FROZEN PLASMA

Volume 43, September 2003

TRANSFUSION 1323

than single stranded, although virusescontaining genomes of either type maybe efficiently inactivated (see below).Activation results in a mixture of strandcross-linking, guanosine oxidation, anddepurination. MB may also modifyproteins and lipids, the relative ratesdepending on the MB and local oxygenconcentrations. For virus-infected cells,this may be influenced by the reducingand detoxifying mechanisms presentinside the cell. MB is not considereduseful for inactivation of intracellularviruses or to attain bacterial or proto-zoal reduction, although it does entercells.

5-7

Its only application in transfu-sion has been to achieve virus inactiva-tion of plasma, with prior cell removalby filtration or freeze-thaw lysis

8-10

(Fla-ment J, Mohr H, and Walker W, writtencommunication, 2000).

Photodynamic treatment with MBresults in efficient virus inactivation forall lipid-enveloped viruses tested to date,including all those for which the UK andUS currently routinely screen blooddonations, as well as West Nile virus.

3-5,10

The extent of removal for such viruses isusually at least 5 logs, this being true forboth double- and single- stranded RNAand DNA viruses (Table 1). Nonlipid-enveloped viruses show a more diversespectrum of susceptibility, some beingtotally unaffected (EMC, polio, HAV, por-cine parvovirus), whereas others (SV40,HEV models, human parvovirus B19)show reduction factors of 4 logs or more(Table 1). More recently, testing usingPCR methods has shown direct removalof HIV, HBV, HCV, and parvovirus B19reactivity from infected donations,

11-14

the last of thesedemonstrating 4-log reduction by a newly developed B19bioassay on the KU 812 EP 6 cell line (Flament J, Mohr H,and Walker W, written communication, 2000).

Are such reduction factors sufficient to assure that asingle plasma donation, taken during the peak of viremia,is rendered noninfectious? The answer will depend onwhether the donation is also subjected to NAT or serologictesting and on the level of viremia. For most viruses, weknow that the answer is almost certainly yes, but in a fewcases such as parvovirus B19, in which the peak of viremiais around 10

7

genome equivalents per mL, this conclusionis more dubious. However, for viruses of major concern,peak viremia levels are either within the clearance rangeof the system, or screening with assays of high sensitivity

will have ensured that only donations with lower levels ofviremia enter the processing laboratory (handling errorsexcepted). In the pregenome testing era, there was a pos-sible HCV exposure from a unit of MBFFP taken from adonor in the sero-negative window period (Flament J,written communication, March 1998). The patient sero-converted for HCV but remained genome negative. Theprecise events remain unproven, but it is possible that thepatient generated an antibody response against inacti-vated virus.

Although MB and other phenothiazine dyes havebeen suggested as having inhibitory action against trans-missible spongiform encephalopathies,

15

there is noevidence of in vitro inactivation of infectivity at the con-centrations used in the transfusion setting.

Fig. 1.

MB and its photodegradation products.

TABLE 1. Virus reduction

Lipid enveloped Non-lipid enveloped

Viruslog reduction

factor Viruslog reduction

factorHIV

>

5.5 HAV 0.0Bovine viral diarrhea

>

6.2 Encephalomyocarditis 0.0Duck HBV 3.9 Porcine parvovirus 0.0Influenza 5.1 Polio 0.0Pseudorabies 5.4 SV40 4.3Herpes simplex

>

6.5 Adenovirus 4.0Vesicular stomatitis

>

4.9 Human parvovirus B19

≥

4.0West Nile virus

>

6.5 Calicivirus (HEV)

>

3.9

16

WILLIAMSON ET AL.

1324 TRANSFUSION

Volume 43, September 2003

EFFECT OF MB TREATMENT ON COAGULATION PROTEINS

It is well established that MB treatment of plasma affectsthe functional activity of various coagulation proteins andinhibitors (Table 2). The proteins most severely affected byMB treatment of plasma are FVIII and fibrinogen, whereactivity is reduced by 20 to 35 percent. The decrease infibrinogen is seen when assayed by the method of Clauss,but not in antigenic assays,

16

suggesting that MB treat-ment effects the biologic activity but not concentration offibrinogen. It has been suggested that this is due to thephoto-oxidation of fibrinogen inhibiting polymerizationof fibrin monomers.

17

The effects on fibrinogen are prob-ably due to an interaction of MB with histidine residuesand may result in a modified in vivo clearance.

16,18-20

How-ever, fibrinogen isolated from MB-treated plasma retainsnormal ability to bind to glycoprotein IIb/IIIa receptors onplatelets,

21

an important mechanism in platelet activation

and aggregation. The inhibitory effects are ameliorated bythe presence of ascorbate

22

but do not appear to result inthe formation of any neoantigens

16,18,19

or positivity in testsfor the formation of IgE antibodies (Flament J, Mohr H,and Walker W, personal communication, 2000).

Unsurprisingly, the changes in coagulation proteinsobserved in MB-treated plasma are associated with a pro-longation of the prothrombin time (PT) and activated par-tial thromboplastin time (APTT).

16,23

Original studies on MB inactivation were reported onplasma freeze-thawed before treatment, but later work onthe Baxter Pathinact and Maco Pharma Theraflex systemswas performed on fresh plasma (Table 2). However, wehave recently shown that the major cause of coagulationfactor loss is the MB treatment itself and not the freeze-thawing.

16,24

Fortunately, changes in coagulation proteinsinduced by WBC-reduction and MB-removal filters appearto be negligible compared to the effect of the MB processitself. Filtration of plasma using a filter (Hemasure)

Fig. 2.

Schematic representation of the closed bag system for MB treatment of fresh-frozen plasma.

The MACO PHARMA Plasma Membrane filtrationThe MACO PHARMA Plasma Membrane filtrationMethyleneMethylene Blue Illumination and MB Depletion SetBlue Illumination and MB Depletion Set

Filtre plasma

PLAS4

Methylen bluedry pill

Plasmaflex PLAS 4Membrane plasma filter

BlueflexMethylene bluedepletion filter

Illuminationbag

Plasmastorage

bag

17

METHYLENE BLUE-TREATED FRESH-FROZEN PLASMA

Volume 43, September 2003

TRANSFUSION 1325

designed to remove both WBCs and MB simultaneouslyresults in a prolongation of the APTT but has no effect onthe PT or fibrinogen when measured by manual tech-niques.

25

Filters to remove residual MB in plasma devel-oped more recently by Pall and Maco Pharma are reportedto result in a small increase in the APTT but minimal lossof coagulation factor activity.

26,27

It has been suggested thatthe increase in the APTT in the latter studies may be aresult of some activation of the contact system of coagu-lation after contact of plasma with the artificial surface ofthe filter.

26

Levels of thrombin-antithrombin complexes are notelevated in MB-treated plasma,

16

indicating that MBtreatment is also not associated with excessive thrombingeneration. Functional measurements of the naturallyoccurring anticoagulants protein C & S and antithrombinalso appear to be relatively unaltered in MB-treatedplasma.

10,16,23,28

MB treatment is reported to have littleeffect on levels of plasminogen, alpha-2-antiplasmin (themain inhibitor of plasmin), fibrin monomer, and

D

-dimers,

16

suggesting that the use of MBFFP is unlikely toresult in enhanced fibrinolysis. vWF activity in plasma, asmeasured by ristocetin-induced agglutination of platelets,is reduced by 10 to 20 percent,

23,29

but vWF multimericdistribution and cleaving protease activity are reported tobe unaffected.

23,28-30

After transfusion of MB-treated plasma to healthyadults, there was no significant difference from baselinevalues in APTT, PT, TT, FVIII, FXI, Clauss fibrinogen, fibrindegradation components, or platelet aggregation induced

by collagen or ADP, suggesting no major influence oncoagulation or fibrinolytic systems.

31

There have been relatively few studies examiningcryoprecipitate and cryosupernatant produced from MBplasma. Levels of FVIII and fibrinogen activity in cryopre-cipitate are 20 to 40 percent lower than untreated units

23,32

but remain within Council of Europe Guidelines. Theeffect on levels of vWF antigen and activity seem morevariable: one study reports no significant difference,

23

whereas in a two-center study, one center also reported nochange, while the other saw 15 to 20 percent lower valuesin MB units.

32

These differences might be explained byvariation in the methodology used to prepare the cryopre-cipitate. However, both studies show that the multimericdistribution of vWF is unaltered. Cryoprecipitate pro-duced from MBFFP has not yet been introduced in anycountry that provides MBFFP, but work is ongoing in theUK to optimize fibrinogen concentration.

33

Cryosupernatant produced from standard or MB-treated plasma lacks the largest molecular weight forms ofvWF.

23

The main clinical indication for cryosupernatant isfor the treatment of thrombotic thrombocytopenic pur-pura (TTP). Patients with TTP tend to have unusually largemolecular weight vWF multimers,

34

which are known topromote platelet aggregation, and some believe that treat-ment with a plasma component that lacks the high molec-ular weight forms of vWF may be beneficial. However, noclinical data are available to answer this question. Levelsof vWF cleaving protease have not been measured in cryo-supernatant produced from MB-treated plasma, but given

TABLE 2. Changes in coagulation factor proteins and inhibitors in MBFFP

Parameter* Percent change due to MB treatment†‡§ Mean residual levels‡§Fibrinogen (Clauss) g/L

Ø

24,

10

24,

23

39

29

1.65,

10

1.80,

16

2.01,

23

1.97,

28

2.05

29

Fibrinogen (antigen) g/L 2.74

16

Prothrombin (FII) (U/mL)

Ø

8,

10

8,

16

18,

23

1.15,

10

1.05,

16

1.00

23

FV (U/mL)

Ø

4.5,

10

21,

16

32,

23

10,

28

0.84,

10

0.73,

16

0.79,

23

0.76

28

FVII (U/mL)

Ø

8,

10

9,

16

7,

23

1.10,

10

0.90,

16

0.90

23

FVIII (U/mL)

Ø

13,

10

33,

16

28,

23

26,

28

29

29

0.78,

10

0.58,

16

0.58,

23

0.83

28

FIX (U/mL)

Ø

17,

10

23,

23

11

28

1.00,

10

0.72,

23

0.88

28

FX (U/mL)

Ø

13,

10

7

23

1.05,

10

0.90

23

FXI (U/mL)

Ø

17,

10

27,

23

13

28

1.00,

10

0.73,

23

0.84

28

FXII (U/mL)

Ø

17

10

1.20

10

FXIII (U/mL)

Ø

7,

23

16

29

1.02,

23

1.12

29

vWF antigen (U/mL)

Ø

7,

23

5

29

Æ

28

0.94,

23

0.83,

29

1.00

28

vWF:ristocetin cofactor(U/mL)

Ø

8,

23

18

29

0.92,

23

0.79

29

C1-inhibitor (U/mL)

Ø

23,

10

Æ

16

0.88,

10

1.03

16

Antithrombin (U/mL)

Ø

8,

10

3

23

Æ

16,23

0.78,10 0.95,16 1.00,23 0.9628

Protein C (U/mL) Æ16,28 1.03,16 0.8928

Protein S (U/mL) Æ16 1.1116

a1-antitrypsin (U/mL) Æ16 155 mg/dLPlasminogen (U/mL) Æ10,16 0.90,10 0.9816

a2-antiplasmin (U/mL) Æ16 0.9616

* Results given as U/mL because not all studies were calibrated against international standards. Assays are functional unless otherwise stated.† Arrows indicate direction of change, with horizontal arrow indicating no change.‡ 10,16,23Studies used frozen-thawed plasma.§ 28,29Studies used fresh plasma (<8 hr from collection).

18

WILLIAMSON ET AL.

1326 TRANSFUSION Volume 43, September 2003

that levels appear to be relatively unaltered in the sourceplasma,30 one would not expect them to differ signifi-cantly. It would thus appear that MB-treated cryosuperna-tant would be suitable for the treatment of TTP, but it hasyet not been manufactured for clinical use.

If MB plasma is used to suspend single-donor plate-lets, there is no significant effect on platelet numbers,morphology scores, osmotic recovery, or levels of LDH,CD62P expression, lacate, pH, and glucose compared tostandard plasma.35 Similarly, if MB-treated plasma isadded to RBCs, there appears to be no appreciable effecton leakage of potassium, hemolysis, or osmotic fragilityduring 28 days of storage.35 This is in contrast to directtreatment of RBCs with MB and light, which results inmembrane leakage and enhanced surface binding ofIgG.5-7

PHARMACOLOGY AND TOXICOLOGY

The major clinical application of MB in the past has beenas a redox reagent in the reversal of methemoglobinemiaand cyanide poisoning using intravenous doses of 1 to5 mg per kg. It has also been used at higher oral doses forthe treatment of manic depression (300 mg/day) and renalcalculus disease (195 mg/day). Intravenous doses of 2 to5 mg per kg have also been used for heparin neutrali-zation and for perioperative staining of the parathyroidgland.3,4,9,10,20 For comparison, the plasma pathogen-reduction systems described here result in a MB concen-tration of 1 mM in the FFP, equivalent to an intravenousdose per 250 mL FFP unit of 0.0012 mg per kg. If MB-removal filters are used during processing,25,36 this level isreduced approximately ¥10, to a final concentration of 0.1to 0.3 mM. For a 70-kg adult receiving the recommended15 mL per kg of FFP, this equates to a total MB dose ofapproximately 33 mg, or less than 1 mg in a 2-kg prematureinfant. Infused MB is rapidly cleared from the circulationand marrow (half-lives in rats are 7 and 18 min, respec-tively) to an extent that its presence in blood (half-life inman approx. 60 min) is difficult to detect after infusion ofMBFFP. There is some tissue uptake, but the majority ofMB is excreted via the gastrointestinal tract and in urinewithin 2 or 3 days3 (Flament J, Mohr H, and Walker W,written communication, 2000).

A US toxicologic report summarizes its use to assessmembrane rupture during amniocentesis, noting mildand transient side effects at most.37 In mammals, the halflethal dose for MB is of the order of 100 mg per kg, withphoto-illumination products having similar, or lesser, tox-icity profiles to the parent compound.3,5 Chronic dosing ofanimals with MB at doses up to 0.2 g per kg day for 13weeks are nontoxic. Chronic exposure of rats to a diet con-taining 4 percent MB had no carcinogenic or cirrhoticeffects, while testing in both rodents and Drosophilarevealed no genotoxic effects at near lethal doses. Testing

for induction of birth defects at doses up to 5 mg per kgper day has also given negative results,3,5 although recentlyhigher doses have been reported as inducing fetal growthretardation.38 In contrast to this, in vitro tests, such as theAmes test for mutagenic effect in selected bacteria, haveyielded some mutagenic and genotoxic data, particularlyin the presence of a liver microsomal (S9) fraction. Testingon human lymphocytes and the mammalian V79 cell linehas been reported by some to show no mutagenicity,although in the presence of the microsomal S9 fraction,some chromosomal aberrations were seen in lymphocytesat 1 to 2 mg per mL (Flament J, Mohr H, and Walker W,written communication, 2000). Wagner et al.39 hasreported genotoxic effects in mouse lymphoma cells at30 mg per mL of MB, which was enhanced by S9 addition,but failed to detect any activity in vivo in a mouse micro-nucleus assay.

Between 1992 and 1998, more than a million units ofMBFFP were used in Germany, Switzerland, Austria, andDenmark. Use has continued in the UK, Portugal, andSpain using the Grifols, Baxter, and Macopharma versionsof the technology. The latter two systems have a EuropeanMedical Devices licence (CE mark), granting of whichincludes a toxicologic assessment. Both passive and activesurveillance40 have yielded adverse event rates that do notdiffer from those for standard FFP. In neonates, where theconcern is greater due to the immature detoxification sys-tem, there are few reports on surveillance, but data fromboth Germany and Spain indicate no acute adverseevents, even when MBFFP is used for exchange transfu-sion (Castrillo A, Pohl U, written communication, 1999).Concern over the potential in vitro mutagenic effects ofMB and its derivatives, particularly in the presence of theS9 fraction, was the reason for the failure to re-license theproduct (without MB removal) in Germany in 1998. Anopinion has not been reached on whether the systemincluding the MB-removal step will be granted a Germanlicense. However, a large amount of clinical usage and invivo toxicology testing suggest that despite the effects seenin vitro, in vivo side effects are minimal, presumablymainly due to the dilution on infusion and the rapid clear-ance of the compound. One toxicology expert in the fieldhas suggested the risk is on a par with smoking a pack ofcigarettes over a lifetime (Flament J, Mohr H, and WalkerW, written communication, 2000).

CLINICAL STUDIES

Most studies in patients have been small and/or have usedlaboratory rather than clinical endpoints. Despite usage ofmore than 1 million units in Europe, there have been nofull reports of large, randomized trials of MBFFP usingrelevant endpoints such as blood loss or exposure to otherblood components. Early studies described successful useof MBFFP in either single or small groups of patients with

19

METHYLENE BLUE-TREATED FRESH-FROZEN PLASMA

Volume 43, September 2003 TRANSFUSION 1327

deficiencies of FV or FXI, TTP, and exchange transfusionin neonates.41,42 One study of 71 patients comparedMBFFP with S/D-treated FFP in cardiac surgery andshowed better replacement of protein S and alpha2-antiplasmin with MBFFP but no difference in bloodloss.20 However, one hospital in Spain has reported thatafter a total switch to MBFFP, FFP demand rose by 56 per-cent, with a two- to three-fold increase in demand forcryoprecipitate, which was not MB treated.43 The authorssuggest that the increase in demand, particularly for cryo-precipitate, may have been required to offset the reducedfibrinogen level in the component. Indeed, after orthope-dic surgery, transfusion of MBFFP has been associatedwith increased reptilase clotting times and ratio of immu-nologic to functional measured fibrinogen,44 suggestingthat MB may interfere with fibrin polymerization in vivo.However, the data from the Spanish study need to be inter-preted with care. In the period studied, which spannedintroduction of MBFFP, 2967 patients received no fewerthan 27,434 units of plasma, but only 24,607 units of RBCs,with 26 percent of admissions receiving FFP only. The veryhigh FFP to RBC ratio (1.11) contrasts sharply with therecent corresponding figure for the UK Transfusion Ser-vices (0.14).45 This suggests very different prescribing prac-tices for FFP between Spain and the UK, including routineuse of FFP in all cardiac surgery procedures in Spain.43

Nevertheless, their study emphasizes the importance ofmonitoring clinical demand after any change to MBFFP, tosee whether the in vitro effects truly result in a require-ment for larger doses.

No specific data are available from studies in neo-nates, but no specific problems have been found. The onlyreport of MB toxicity in a neonate was a case of severebullus formation and desquamation was reported in ababy who received phototherapy for hyperbilirubinemiaafter administration of 10 mL of 1 percent MB to themother to investigate possible rupture of amniotic mem-branes.46 Although neonatal blood levels of MB were notreported, the skin of the baby was visibly stained blue,suggesting blood and tissue levels many times higher thanwould be achieved after infusion of MBFFP. No problemswith MBFFP-treated infants requiring phototherapy havebeen reported in Europe, and glucose 6 phosphate dehy-drogenase deficiency is not a contra-indication to its use(Walker W, written communication, 2002). Similarly, digi-tal capillary measurement of oxygen saturation by colori-metric means is not affected by infusion of MBFFP.

Limited data are available on the use of MBFFP forplasma-exchange procedures for TTP.47 Although levels ofvWF cleaving enzyme in MBFFP are normal,30 one studyof two small cohorts of patients (13 treated with FFP and7 with MBFFP) reported an increase in the number ofplasma-exchange procedures and days in hospital in theMBFFP group.48 This is of concern, although the smallpatient numbers make it difficult to draw conclusions;

clearly, larger studies are required to establish the role ofMBFFP in TTP.

FFP SAFETY: WHERE ARE WE GOING?

Five years ago, an editorial in this journal accompaniedthe availability in the USA of pooled S/D FFP.1 Despite theimpact of the previous HIV and HCV transmissions ontransfusion services in many countries, S/D FFP did notsubsequently become a standard of care in the USA,although it has become so in Norway, Belgium and Portu-gal. Other European countries have chosen quarantiningof FFP with donor re-test as their method of minimizingvirus risk from FFP. This avoids potential toxicity or loss ofactivity, but provides no protection against new agentssuch as West Nile virus. In virus reduction terms, the MBsystem appears to have acceptable efficacy, and has theadvantage of being a single unit system, so that potentiallyincreased risks from new agents unaffected by the system,such as prions, are minimized. The major disadvantage isloss of coagulation factors, such as fibrinogen. The as yetunlicenced single unit psoralen S59 pathogen reductionsystem for FFP appears to result in much better preserva-tion of fibrinogen, with only 3-13 percent reduction.49

However, toxicity will be a concern for any pathogenreduction system which interacts with nucleic acids, espe-cially if administered to very young recipients.

In the UK, provision of MBFFP is linked to the mostrecent Department of Health precautionary decision tominimize the unknown risk of variant CJD from UK bloodcomponents. In August 2002, UK Transfusion Serviceswere instructed to seek supplies of US plasma for FFPproduction for children born after January 1, 1996, a datefrom which the UK food supply has been considered safefrom bovine spongiform encephalopathy. This importedFFP will be subjected to MB treatment, and, in prepara-tion, UK Transfusion Services have already introducedMBFFP for this age group. No immediate problems withside effects or loss of efficacy have been reported,although the number of children treated is still small.Hospitals also have access to S/D FFP from commercialsources.

But to take an overview of FFP safety, 5 years’ hemo-vigilance data in the UK reveal that virus transmission is amuch smaller risk than that of TRALI. From 1996 to 2001,there were 15 TRALI cases in which FFP was clearly impli-cated, and another 4 where FFP was among a range ofcomponents transfused. In the same time period, therewas not a single proven virus transmission from FFP.45

Single-unit pathogen-reduction systems by themselvescontribute nothing to TRALI prevention, which may behelped by selection of male donors for FFP50 and/orscreening of parous females for WBC antibodies. Interest-ingly, the pooling of several hundred donations requiredin the S/D FFP process may provide benefit against TRALI

20

WILLIAMSON ET AL.

1328 TRANSFUSION Volume 43, September 2003

by diluting out those with high-titer WBC antibodies. TheNational Blood Service in England has begun a formaloption appraisal of TRALI-prevention strategies, begin-ning with plasma-rich components. The relative cost effec-tiveness and long-term role of pathogen reduction of FFPin an overall blood safety strategy remain to be elucidated.

REFERENCES

1. Klein HG, Dodd RY, Dzik WH, et al. Current status of solvent/

detergent-treated frozen plasma. Transfusion 1998;38:102-7.

2. Provisional surveillance summary of the West Nile Virus

epidemic—United States January-November, 2002. MMWR

2002;51:1129-33.

3. Chapman J. Virus inactivation of plasma by methylene blue

photoinactivation: approaches and toxicology. Transfus

Today 1994;20:2-4.

4. Mohr H, Lambrecht B, Selz A. Photodynamic virus

inactivation of blood components. Immunol Invest

1995;24:73-85.

5. Wagner SJ. Virus inactivation in blood components by

photoactive phenothiazine dyes. Transfus Med Rev

2002;16:61-6.

6. Wagner S, Storry JR, Mallory DA, et al. Red cell alterations

associated with virucidal methylene blue treatment.

Transfusion 1993;33:30-6.

7. Wagner SJ, Robinette D, Storry J, et al. Differential

sensitivities of viruses in red cell suspensions to methylene

blue photosensitization. Transfusion 1994;34:521-6.

8. Abe H, Yamada-Ohnishi Y, Hirayama J, et al. Elimination of

both cell-free and cell-associated HIV infectivity in plasma

by a filtration/methylene blue photoinactivation system.

Transfusion 2000;40:1081-7.

9. Chapman J. Viral inactivation of plasma by methylene blue

photoinactivation: virology and plasma protein overview.

Transfus Today 1994;22:2-5.

10. Lambrecht B, Mohr H, Knuver-Hopf J, Schmitt H.

Photoinactivation of viruses in human fresh plasma by

phenothiazine dyes in combination with visible light. Vox

Sang 1991;60:207-13.

11. D’Antonio D, Di Gianfilippo R, Iacone A, et al. Methylene

blue and high intensity light source inducing HCV RNA PCR

negativity in plasma of two HCV positive patients.

Transfusion 1999;39:24S.

12. Mohr H, Bachmann B, Klein-Struckmeier A, Lambrecht B.

Virus inactivation of blood products by phenothiazine dyes

and light. Photochem Photobiol 1997;65:441-5.

13. Muller-Breitkreutz K, Mohr H. Hepatitis C and human

immunodeficiency virus RNA degradation by methylene

blue/light treatment of human plasma. J Med Virol

1998;56:239-45.

14. Iudicone P. Photodynamic treatment of fresh frozen plasma

by methylene blue: effect on HIV, HCV and parvovirus B19.

Infusion Ther Transfus Med 1999;26:262-6.

15. Achour A. Phenothiazines and prion diseases: a potential

mechanism of action towards oxidative stress. Int J

Antimicrob Agents 2002;20:305-6.

16. Zeiler T, Riess H, Wittmann G, et al. The effect of methylene

blue phototreatment on plasma proteins and in vitro

coagulation capability of single-donor fresh-frozen plasma.

Transfusion 1994;34:685-9.

17. Inada Y, Hessel B, Blomback B. Photooxidation of fibrinogen

in the presence of methylene blue and its effect on

polymerization. Biochim Biophys Acta 1978;532:161-70.

18. Mohr H, Knuver-Hopf J, Lambrecht B, et al. No evidence for

neoantigens in human plasma after photochemical virus

inactivation. Ann Haematol 1992;65:224-8.

19. Tissot U, Hochstrasser D, Schneider B, et al. No evidence for

protein modifications in fresh frozen plasma after

photochemical treatment: an analysis by high-resolution

two-dimensional electrophoresis. Br J Haematol 1994;

86:143-6.

20. Wieding JU, Rathgeber J, Zenjer D. Prospective, randomized

trial and controlled study on solvent detergent versus

methylene blue virus inactivated plasma. Transfusion

1999;39:23S.

21. Lorenz M, Muller M, Jablonka B, et al. High doses of

methylene blue/light treatment crosslink the A-alpha-

subunit of fibrinogen: influence of this photooxidization on

fibrinogen binding to platelets. Haemostasis 1998;28:17-24.

22. Parkkinen J, Vaaranen O, Vahtera E. Plasma ascorbate

protects coagulation factors against photooxidation.

Thromb Haemost 1996;75:292-7.

23. Aznar JA, Bonanad S, Montoro JM, et al. Influence of

methylene blue photoinactivation treatment on coagulation

factors from fresh frozen plasma, cryoprecipitates and

cryosupernatants. Vox Sang 2000;79:156-60.

24. Cardigan R, Garwood M, Hornsey V, et al. Effect of freeze-

thawing, leucocyte depletion, methylene blue (MB)

treatment and removal on the quality of FFP. Vox Sang

2002;83:190.

25. AuBuchon JP, Pickard C, Herschel L, et al. Removal of

methylene blue from plasma via an adsorbent filter. Vox

Sang 1998;74:1-6.

26. Riggert J, Humpe A, Legler TJ, et al. Filtration of methylene

blue-photooxidized plasma: influence on coagulation and

cellular contamination. Transfusion 2001;41:82-6.

27. Verpoort T, Chollet S, Lebrun F, et al. Elimination of

methylene blue from photo-dynamically treated virus

inactivated fresh-frozen plasma: the Blueflex filter. Trans

Clinique Biologique 2001;8(Suppl 1):103s.

28. Hornsey VS, Drummond O, Young D, et al. A potentially

improved approach to methylene blue virus inactivation of

plasma: the Maco Pharma Maco-Tronic system. Transfus

Med 2001;11:31-6.

29. Aznar JA, Molina R, Montoro JM. Factor VIII/von Willebrand

factor complex in methylene blue-treated fresh plasma.

Transfusion 1999;39:748-50.

30. Cardigan R, Allford S, Williamson L. Levels of von Willebrand

21

METHYLENE BLUE-TREATED FRESH-FROZEN PLASMA

Volume 43, September 2003 TRANSFUSION 1329

factor cleaving protease are normal in methylene blue

treated fresh frozen plasma. Br J Haematol 2002;117:253-4.

31. Simonsen AC, Sorensen H. Clinical tolerance of methylene

blue virus-inactivated plasma. Vox Sang 1999;77:210-7.

32. Hornsey VS, Krailadsiri P, MacDonald S, et al. Coagulation

factor content of cryoprecipitate prepared from methylene

blue plus light virus-inactivated plasma. Br J Haematol

2000;109:665-70.

33. Prowse CV, Hornsey V, Young D. Improving the yield of

fibrinogen in cryoprecipitate prepared from methylene blue

treated (MB) fresh frozen plasma. International Society of

Blood Transfusion, Istanbul, 2003. Abstract 025.

34. Moake JL, McPherson PD. Abnormalities of von Willebrand

factor multimers in thrombotic thrombocytopenic purpura

and the hemolytic-uremic syndrome. Am J Med 1989;87:9N-

15N.

35. Perrotta PL, Baril L, Tead C, et al. Effects of methylene blue-

treated plasma on red cells and stored platelet concentrates.

Transfusion 1999;39:63-9.

36. Garwood M, Cardigan R, Drummond O, et al. The effect of

methylene blue photoinactivation and methylene blue

removal on the quality of fresh-frozen plasma (FFP).

Transfusion 2003;43:1238–47.

37. Little AD Inc. National Toxicology Program. Executive

summary of safety and toxicity information: methylene

blue. Arthur D. Little, Cambridge, MA, 1990.

38. Tiboni GM, Giampetro F, Lamonaca D. The soluble

guanylate cyclase inhibitor methylene blue evokes pre-term

delivery and fetal growth retardation in a mouse model. In

Vivo 2001;15:333-7.

39. Wagner SJ, Cifone MA, Murli H, et al. Mammalian

genotoxicity assessment of methylene blue in plasma:

implications for virus inactivation. Transfusion 1995;35:

407-15.

40. Pohl U, Mohr H. Virus Inactivation of plasma by methylene

blue/light exposure: clinical experiences. Transfus Today

1994;21:2-3.

41. Pohl U, Wieding JU, et al. Treatment of patients with severe

congenital coagulation factor V or Xi deficiency with

methylene blue virus inactivated plasma. ISBT 5th Regional

(European) Congress, Meeting Proceedings, Venice, Italy,

1995.

42. Pohl U, Becker M, et al. Methylene blue virus inactivated

plasma in the treatment of patients with thrombotic

thrombocytopenic purpura. ISBT Vth Regional (European)

congress, Meeting Proceedings, Venice, Italy, 1995.

43. Atance R, Pereira A, Ramirez B. Transfusing methylene blue-

photoinactivated plasma instead of FFP is associated with

an increased demand for plasma and cryoprecipitate.

Transfusion 2001;41:1548-52.

44. Taborski U, Oprean N, Tessmann R, et al. Methylene blue/

light-treated plasma and the disturbance of fibrin

polymerization in vitro and in vivo: a randomized, double-

blind clinical trial. Vox Sang 2000;78:P546.

45. Serious Hazards of Transfusion Annual Report, 2000-2001.

SHOT Steering Committee, 2002.

46. Porat R, Gilbert S, Magilner D. Methylene blue-induced

phototoxicity: an unrecognized complication. Pediatrics

1996;97:717-21.

47. Pohl U, Becker B, Papstein C, et al. Methylene blue virus

inactivated fresh frozen plasma in the treatment of patients

with thrombotic thrombocytopenic purpura. Meeting

Proceedings, Japan: International Society of Blood

Transfusion, 1996.

48. De la Rubia J, Arriaga F, Linares D, et al. Role of methylene

blue-treated or fresh-frozen plasma in the response to

plasma exchange in patients with thrombotic

thrombocytopic purpura. Br J Haematol 2001;114:

721-3.

49. Alfonso R, Lin C, Dupuis K, et al. Inactivation of viruses with

preservation of coagulation function in fresh frozen plasma.

Blood 1996;88:10, S1, 526a.

50. Palfi M, Berg S, Ernerudh J, Berlin G. A randomized

controlled trial of transfusion-related acute lung injury: is

plasma from multiparous blood donors dangerous?

Transfusion 2001;41:317-22.

51. Lillie RD. Conn’s biological stains. Baltimore: Williams and

Wilkins, 1991.

52. Budavari S (ed). The Merck index, 12th edn. Whitehouse

Station (NJ): Merck and Co., 1996.

22

1238 TRANSFUSION

Volume 43, September 2003

Blackwell Science, LtdOxford, UKTRFTransfusion0041-11322003 American Association of Blood BanksSeptember 200343Original Article

METHYLENE BLUE PHOTOINACTIVATION OF PLASMAGARWOOD ET AL.

ABBREVIATIONS:

APC

=

allophycocyanin-conjugated; APTT

=

activated partial thromboplastin time; FFP

=

fresh-frozen plasma;

MB

=

methylene blue; PMN,

=

neutrophil; PRP

=

platelet-rich

plasma; PT

=

prothrombin time; VWF:CB

=

VWF collagen-binding

activity.

From the National Blood Service, England & North Wales; the

Scottish National Blood Transfusion Service, Edinburgh and

Glasgow; and the University of Cambridge, United Kingdom

Address reprint requests to:

Rebecca Cardigan, PhD, National

Blood Service, Crescent Drive, Brentwood, Essex CM15 8DP, UK;

e-mail: rebecca.cardigan@nbs.nhs.uk.

This work was partly funded by Maco Pharma, Middlesex,

UK.

Received for publication January 29, 2003; revision received

April 7, 2003, and accepted April 22, 2003.

TRANSFUSION

2003;43:1238-1247.

B L O O D C O M P O N E N T S

The effect of methylene blue photoinactivation and methylene blue removal on the quality of fresh-frozen plasma

Margaret Garwood, Rebecca A. Cardigan, Olive Drummond, Valerie S. Hornsey, Craig P. Turner,

David Young, Lorna M. Williamson, and Chris V. Prowse

BACKGROUND:

The effects of using fresh or frozen-thawed plasma, WBC reduction of plasma before freezing, and the use of two different methylene blue (MB) removal filters on the quality of MB-treated plasma were compared.

STUDY DESIGN AND METHODS:

In a paired study (n

=

11/arm) plasma was frozen within 8 hours of collection, thawed, MB photoinactivated, and then filtered using one of two MB removal filters. Fresh plasma (n

=

16) and plasma WBC reduced before freezing (n

=

19) were MB inactivated.

RESULTS:

Freeze-thawing resulted in loss of activity of FXII and VWF of 0.06 and 0.04 units per mL, respectively, but no significant loss of activity of factors II through XI or fibrinogen. Further loss of activity occurred after MB treatment: FII (0.07 IU/mL), FV (0.11 U/mL), FVII (0.08 IU/mL), FVIII (0.28 IU/mL), F IX (0.12 IU/mL), FX (0.16 IU/mL), FXI (0.28 U/mL), FXII (0.15 U/mL), VWF antigen (0.05 IU/mL), VWF activity (0.06 U/mL), and fibrinogen (0.79 g/L). Losses due to this step were significantly (5-10%) lower in fresh plasma compared to frozen-thawed plasma. Neither MB removal filter resulted in significant loss of activity of any factor studied.

CONCLUSION:

MB removal, by either of the available filters, has little impact on the coagulation factor content of plasma, but freezing of plasma before MB treatment results in a small additional loss.

ue to stringent donor selection and testingprocedures, fresh-frozen plasma (FFP) in thedeveloped world offers a high degree of viralsafety. For example, the risk of an infectious

FFP donation entering the blood supply in England is esti-mated to be 1 in 10 million for HIV, 1 in 50 million for HCV,and 1 in 1.2 million for HBV (Eglin R, written communica-tion, 2002). Nevertheless, viral transmission from bloodcomponents continues to occur, with 16 cases reported inthe UK in the last 6 years.

1

There is, therefore, considerableresearch activity in pathogen inactivation of single-unitcomponents because methods suitable for single compo-nents offer reassurance that no increased infectious risksare added due to pooling. For FFP only, one licensedsingle-unit system is currently available (methylene bluephotoinactivation). It is desirable that there is as muchflexibility as possible in the handling conditions forplasma before inactivation, to enable production of FFPfrom collection centers distant from the processing site.This is particularly relevant because the UK Departmentsof Health have recently recommended that FFP isimported from North America for neonates and childrenborn after 1995 (after the introduction of relevant foodbans to limit BSE transmission) as a precautionary mea-sure against vCJD transmission. Previous studies have

D

23

METHYLENE BLUE PHOTOINACTIVATION OF PLASMA

Volume 43, September 2003

TRANSFUSION 1239

demonstrated vCJD infectivity in plasma of rodentsinfected with prion diseases,

2,3

and a recent reportdescribes interim results from a study that demonstratetransmissions of bovine spongiform encephalopathy(BSE) and scrapie between sheep by whole-blood transfu-sion.

4

Because background levels of virus marker positivityin the North American population are significantly higherthan in the UK, it has been deemed sensible to subjectimported plasma to a pathogen-inactivation step.

The methylene blue (MB) photoinactivation processfor viral inactivation of human plasma has been welldescribed

5

as has its effect on the loss of coagulation factoractivity of plasma.

5-9

The original Springe MB process, alsoused by Grifols in Spain, described by Lambrecht,

5

usedfreeze-thawing of plasma before MB inactivation toexpose intracellular viruses to the action of MB. However,recently, blood collection packs that integrate WBC reduc-tion and MB addition before inactivation of plasma(Baxter Pathinact, Baxter Healthcare, Compton Newbury,Berkshire, UK, and Maco Pharma Theraflex, Middlesex,UK) remove the need to freeze-thaw plasma.

10-12

There arealso differences between the systems in how MB is addedto plasma. With two of the systems (Springe and Baxter),a variable dose of MB solution is added to achieve a stan-dard final concentration of 1

m

M

MB. The other system(Maco Pharma Theraflex) incorporates an 85-

m

g pellet ofMB hydrochloride per plasma unit, therefore the concen-tration can vary slightly (0.84-1.13

m

M

) depending uponthe plasma volume (recommended range, 235-315 mL).

For MB photoinactivation of plasma to be centralized,but plasma from remote sites used as a start material, it isessential to be able to freeze and thaw plasma before treat-ment. Although we have previously evaluated the use ofthe two systems (Baxter and Maco Pharma) using freshplasma,

10-12

we have not evaluated freeze-thawing ofplasma before MB treatment using such systems. Althoughit is known that freeze-thawing itself has minimal effect onthe coagulation factor activity of plasma,

7

there are nocomparative data available on whether the loss of coagu-lation factor activity due to the MB inactivation step isaffected by prior freeze-thawing. Furthermore, in the UK,there was concern that freeze-thawing non-WBC-reducedplasma could potentially increase exposure to vCJD due tofragmentation of platelets and WBCs, which are known tocontain normal cellular prion protein (Prp

c

)

13

and mighttherefore host the infective abnormal prion protein Prp

sc

.We therefore assessed the effect of removing these cells byan additional WBC reduction step before freezing on thequality of MB plasma.

Following concerns over possible side effects ofresidual MB in plasma, a further recent development isthe ability to remove MB by filtration before final com-ponent storage. Evaluations of two removal filters (PallMB1, Pall Biomedical, Portsmouth, UK, and HemaSureLeukoVir, Marlborough, MA) MB have been previously

reported,

14,15

but there are no data available on the use ofa new MB removal filter (Maco Pharma Blueflex). The aimof this study was therefore to evaluate the combinedeffect of WBC reduction before freezing, freeze-thawing,MB photoinactivation, and MB removal using two differ-ent filters, on coagulation factor activity and activationmarkers in FFP. We also examined the effect of freeze-thawing and subsequent filtration of non-WBC-reducedplasma on its cellular constituents, to provide assurancethat the process is not likely to increase the risk of vCJDtransmission after transfusion to patients.

MATERIALS AND METHODS

Blood collection and processing

The experimental design is shown in Fig. 1. Twenty-fourunits of whole blood (group A, n

=

12; group O, n

=

12) werecollected into “Top and Bottom” configuration bloodpacks (Pall Medsep 789-94 U, Pall Biomedical). Blood wasthen centrifuged (Heraeus Cryofuge 6000, Kleinostheim,Germany) at 3300 rpm for 12 minutes at 22

∞

C andprocessed to RBCs and plasma (Compomat G4 system,Fresenius-Hemocare NPBI, Abingdon, UK). In ExperimentA, plasmas were pooled in groups of two units of identicalABO group into 600-mL transfer packs (Baxter FGR2089,Baxter Healthcare). The pools were mixed thoroughlyand divided equally between two 300-mL transfer packs(Fresenius Hemocare P4164, Fresenius Hemocare). Allunits of plasma were frozen within 8 hours of collection ina freezer (Thermogenesis MP1101, Cheshire, UK) to

-

45

∞

Cwithin 45 minutes and stored frozen at

-

40

∞

C for 4 days to4 weeks. The units were then thawed at 37

∞

C and immedi-ately WBC reduced (PLAS 4, Maco Pharma, Middlesex, UK)and MB photoinactivated (Maco Pharma Maco-Tronic sys-tem) as previously described.

11

For each pair of plasmas,MB was removed (either Pall MB1 or Maco Pharma Blue-flex) according to the manufacturers’ instructions. Plasmaswere refrozen in a freezer (Thermogenesis MP1101) andstored at

-

40

∞

C. In addition, 19 units of plasma were WBCreduced before freezing using one of two filters (RZ2000,Baxter Healthcare, or LPS1, Pall Biomedical), MB inacti-vated, and then MB removed (MB1 filter, Experiment B).These filters were selected because they are known to haveminimal effect on coagulation factors in plasma.

16

Afurther 16 units of plasma (group O, n

=

8; group A, n

=

8) were MB inactivated without the freeze-thaw step andMB removed (Blueflex filter, Experiment C).

We took 15-mL samples by sterile connection of asample pouch at four time points: 1) before freezing, 2)after thawing and before WBC reduction and MB addition,3) after MB treatment before MB removal, and 4) after MBremoval. Samples were frozen at

-

80

∞

C for coagulationassays, and two aliquots were frozen in EDTA for C3a desarg assays.

24

GARWOOD ET AL.

1240 TRANSFUSION

Volume 43, September 2003

Plasma factors

All coagulation assays were performed using commer-cially available analyzers (Sysmex CA 1500 analyzer,Sysmex, Milton Keynes, UK; Coagamate X2 analyzer,Organon-Teknika, Cambridge, UK; or Amelung KC 4 Amicro analyzer, Sigma Diagnostics

,

Poole, Dorset, UK). FII,FV, FVII, and FX were assayed by one-stage prothrombintime (PT)-based assays and F IX and FXII using a one-stage activated partial thromboplastin time (APTT)-basedassay, using deficient plasma (Dade Behring, Marburg,Germany). The PT and APTT were expressed as a ratio tothe geometric mean result of 20 normal citrated plasmas.These types of samples were chosen as “normal” plasmato provide a standard reference point between studies.VWF antigen was measured by latex agglutination (STALiatest Kit, Diagnostica Stago, Asnieres, France). FVIII andFXI were assayed using one-stage clotting assays with defi-cient plasma (Diagnostics Scotland, Edinburgh, Scotland;and Sigma-Aldrich Company, Poole, Dorset, UK, respec-tively). Fibrinogen was measured using a Clauss assaywith Fibriquick reagents (Organon-Technika, Cambridge,UK). FVIII assays were standardized using the British

plasma standard (NIBSC, South Mimms, UK). All otherassays were standardized using Coagulation Referenceplasma 100 percent (Technoclone, Dorking, UK). A controlplasma of known potency was assayed on each occasionfor all coagulation assays.

Commercially available ELISA kits were used to deter-mine levels of prothrombin fragment 1

+

2 (Dade-Behring), FXIIa (Axis-Shield, Dundee, Scotland), and VWFcollagen-binding activity (VWF:CB, Immuno, Vienna,Austria). C3a des arg was assayed by radioimmunoassay(Amersham Pharmacia Biotech, Buckshire, UK). VWFcleaving protease activity was measured as previouslydescribed

17

and results expressed as a ratio to that of apooled normal citrated plasma.

Effect of freeze-thawing plasma and filtration steps on cellular content of plasma

Double WBC-reduced plasma (LPS1 filter, Pall Biomedi-cal) was spiked with WBCs (

<

1-200

¥

10

6

/L) with or withoutplatelets (

<

1-100

¥

10

9

/L), both of which were preparedfrom fresh whole blood by density gradient centrifugation,

Fig. 1.

Study design.

Experiment A Experiment B Experiment C

2 plasma units Plasma WBC-reduced using either Plasma (<8 hr after collection) pooled and split Baxter RZ2000 filter WBC-reduced using PLAS 4 filter

or Pall LPS1 filter

Frozen < 8 hr after collection Frozen < 8 hr after collection MB added and photoinactivatedStored at –400C Stored at –400C

Thawed at 370C and WBC-reduced 0Thawed at 37 C and WBC-reduced MB removal using PLAS 4 filter using PLAS 4 filter Blueflex filter

n = 16

MB added and photoinactivated MB added and photoinactivated

MB removal MB removal MB removal MB1 filter Blueflex filter MB1 filter n = 12 n = 12 n = 19

25

METHYLENE BLUE PHOTOINACTIVATION OF PLASMA

Volume 43, September 2003

TRANSFUSION 1241

to represent levels of cellular contamination that may beexpected to occur in non-WBC-reduced plasma. Plasmawas then blast-frozen, thawed, WBC-reduced by sterileconnection with the (PLAS 4) filter, MB added, and MBremoved using a filter (the Pall MB1 filter). Samples weretaken at four stages: before freezing, after thawing, afterPLAS4 LD filter, and after MB removal. Samples were ana-lyzed for platelet count by a hematology analyzer (SysmexSE9000, Sysmex) and WBC count by flow cytometry(FACSCalibur, Becton Dickinson, Oxford, UK) usingLeucoCount reagents (Becton Dickinson). Release of theneutrophil primary granule marker elastase was measuredby ELISA of

a

1

-proteinase inhibitor: neutrophil (PMN)elastase complexes (Pathway Diagnostics, UK) and releaseof LDH by enyzymatic assay in supernatant plasma (VitrosDT60II, Axis-Shield, Dundee, Scotland). RBC microparti-cles were measured by flow cytometry (FACSCalibur) aspreviously described using antibodies to glycophorin A.

18

Analysis of platelet microparticles (PMP) was deter-mined as follows: Plasma (5

m

L) was incubated for 20minutes at room temperature with 5

m

L allophycocyanin-conjugated anti-CD61 (APC-CD61, Caltag-Medsystems,Towcester, UK), 5

m

L rhodophycoerythrin-conjugatedanti-CD42b (Caltag-Medsystems), 10

m

L FITC annexin V(FITC-AV, Caltag-Medsystems), 5

m

L 10

¥

HBSS (Sigma,Poole, UK) and made up to 50

m

L with HEPES-calciumbuffer (2.8 m

M

CaCl

2

, 20 m

M

HEPES). Samples were resus-pended in 0.45 mL of 1

¥

HBSS and transferred to a tubecontaining a known amount of beads (Trucount, BectonDickinson), and analyzed using a flow cytometer(FacsCalibur, Becton-Dickinson). Platelet microparticleswere defined using forward scatter as events falling in aregion, which includes less than 2 percent of platelets inplasma (PRP) from 20 normal donors, and of less than1

m

m as determined by APC fluorescent beads (Spherotec,Libertyville, IL). Platelet-derived events were defined byfluorescence due to APC-CD61 binding above that of anisotype-matched control. Annexin-V-positive events weredefined as events binding FITC-AV above a control con-taining 5 m

M

Na

3

EDTA. In normal subjects (n

=

20), lessthan 1 percent of unstimulated platelets bind FITC-AV. Tocontrol assay variability, a negative control of unstimu-lated PRP and a positive control (PRP incubated with10

m

M

A23187 [Calibochem-Novabiochem, San Diego, CA]for 15 min) were included for platelet microparticle assays.

Effect of MB on assays

To assess the effect of the presence of MB itself on coagu-lation factor assays, a MB pellet (from the Maco Pharmapack) was dissolved in each of six units of plasma. Sampleswere collected before and after addition of MB, and onceMB had been added, the plasma was not photoillumi-nated and was protected from light at all times. All pa-rameters were performed as for the main study with the

exception that all coagulation assays were performed us-ing a particular analyzer (Sysmex CA 1500 analyzer) andC3a des arg levels were performed by ELISA (Quidel, SanDiego, CA).

Statistical analysis

Since the distribution of some data were non-Gaussianwith positive skew, nonparametric tests were applied. TheWilcoxon rank sum test was used for paired data and theMann-Whitney U-test for unpaired data. A p value lessthan 0.05 was considered significant. All results are givenas median with range.

RESULTS

Plasma processing

From the 24 paired units (Experiment A), one unit ofplasma fractured on thawing, therefore data on 11 pairedunits of plasma are presented, all of which were within therequired volume range before MB inactivation. Due tosampling, 2 out of 16 fresh plasma units (Experiment C)were slightly below the lower range limit (233 and 234 mL).Filtration time for one WBC-reduction filter (PLAS 4) was8 to 15 minutes, with a loss of 20 mL of plasma. Filtrationtimes for two other removal filters (Pall MB1 and MacoBlueflex) were 2 to 5 minutes and 5 to 11 minutes, respec-tively, with a loss of 10 mL of plasma for each.

Effect of freeze-thawing and MB on loss of coagulation factors

The change in coagulation activity due to freeze-thawingand the MB process is shown in Table 1. There was a sig-nificant loss of FXII (2%) and VWF:CB (9%), and a smallincrease in levels of FVII (1%) and FXI (4%) due to freeze-thawing. This was associated with an increase in both PTratio (1.06 [0.97-1.12] vs. 1.05 [0.95-1.11], p

<

0.0001) andAPTT ratio (1.04 [0.91-1.20] vs. 1.02 [0.89-1.18], p

<

0.0001).The degree of loss of coagulation activity due to the MBprocess varied between factors, the highest losses occur-ring with FVIII (29%), fibrinogen (28%), and FXI (25%),therefore the evaluation of plasma treated fresh wasmainly restricted to these factors. For FV, FVIII, FXI, andfibrinogen, the loss of activity due to the MB process wasapproximately 8-percent higher in the frozen-thawedplasma units compared with fresh plasma (Table 1). Inaddition, the increase due to this step in both PT ratio(0.09 [0.04-0.16] fresh vs. 0.14 [0.08-0.28] frozen-thawed, p

<

0.0001) and APTT ratio (0.11 [0.05-0.15] fresh vs. 0.16[0.11-0.26] frozen-thawed, p

<

0.0001) was also higher.However, there was no significant difference in changes inlevels of FVII, FXIIa, and C3a due to MB treatmentbetween units which were treated fresh or after freeze-

26

GARWOOD ET AL.

1242 TRANSFUSION

Volume 43, September 2003

thaw. The increase in prothrombin

F

1

+

2 levels due to MBtreatment was higher in frozen-thawed units compared tofresh.

When the influence of MB on the assays was studied(in the absence of photoinactivation), there was no signif-icant difference between before or after the addition of MBfor any parameters, apart from FXIIa, which was signifi-cantly lower after MB addition (1.59 [0.76-2.13] ng/mLbefore, 0.72 [0.56-1.05] ng/mL after, p

<

0.05 before vs.after).

Effect of MB removal filters on coagulation activity

To assess the difference between the two MB removal fil-ters, pairs of units were pooled and half of each pool MB-treated and processed through each of the removal filtersin parallel. Due to logistical problems, it was not possibleto process and assay these two sets simultaneously. Asmall difference was apparent in levels of some coagula-tion factors between the two arms of the study, probablydue to small differences in processing and storage. How-ever, the percentage change in activity due to freeze-thawing and MB treatment was the same for each arm ofthe study (data not shown), and therefore these data werecombined. To evaluate the effect of MB removal filters, acomparison of pre- and postcoagulation activity for eacharm of the study was examined (Table 2). There was noapparent decrease in any parameter studied with eitherremoval filter, apart from a reduction in levels of C3a usingone of the filters (Pall MB1). There was a slight increase inlevels of fibrinogen, FII, FV, FVII, and FX, using the Pallfilter. There was an apparent increase in levels of FXIIa

after filtration with both removal filters, which was prob-ably due to the influence of MB on the assay and wascomparable between filters. There was an extremely smallvariation in the PT and APTT ratios with both filters (Table2).

We compared the final levels of coagulation activity infrozen-thawed MB-treated plasma to a reference rangebased on 66 samples of WBC-reduced plasma that had notbeen MB treated. Because there was no loss of activity witheither MB removal filter, both sets of data were pooled.The reference data was not collected as part of this studybut from previous studies carried out by the NationalBlood Service over the past 4 years. The methodology usedin these studies

16

for either plasma processing or assay didnot differ significantly from the current study. Over 90 per-cent of MB units were within our reference range for allcoagulation factors, apart from prothrombin

F

1

+

2 and PTratio, the MB-treated plasma having 23 percent and 50percent of values above the range, respectively (Table 3).The range of PT ratios observed in reference and MB-treated plasma is shown in Fig. 2.

We did not evaluate MB removal by the filters used inthis study, but previous studies have shown that the MB1filter removes 81 to 95 percent of MB

14

and the Blueflexfilter removes more than 95 percent MB.

19

Effect of WBC reduction before freezing

WBC reduction of plasma before freezing appeared tohave little influence on final levels of FVIII (0.67 [0.44-1.23]WBC reduced vs. 0.62 [0.48-0.86 IU/mL] non-WBCreduced) or fibrinogen (1.88 [1.45-3.24] WBC reduced vs.

TABLE 1. Percentage change in coagulation factor activity due to freeze-thawing and MB treatment of plasma

FactorDue to freeze-thawing

median (range)

Due to MB treatment

+

WBC reduction* (freeze-thawed plasma)median (range)

Due to MB treatment

+

WBC reduction (fresh plasma)median (range)

Number 22 22 16Fibrinogen (g/L)

-

10 (

-

27 to 14)

-

28 (

-

51 to

-

20)

-

21 (

-

38 to

-

7)†FII (IU/mL) 0 (

-

4 to 4) –8 (

-

11 to

-

2) NAFV (U/mL) 0 (

-

5 to 4) –13 (

-

20 to 4)

-

5 (

-

11 to 4)†FVII (IU/mL) 1 (

-

1 to 4)||

-

7 (

-

10 to

-

1)

-

4 (

-

9 to

-

1)FVIII (IU/mL)

-

3 (

-

20 to 9)

-

29 (

-

42 to

-9) -24 (-37 to -11)†F IX (IU/mL) 1 (-3 to 4)§ -13 (-20 to -11) NAFX (IU/mL) 0 (-2 to 3) -15 (-22 to -10) NAFXI (U/mL) 4 (-19 to 11)|| -25 (-35 to -7) -15 (-23 to -6)†FXII (U/mL) -2 (-6 to 1)||§ -18 (-31 to -14) NAVWF:Ag (IU/mL) -1 (-4 to 3)§ -6 (-11 to -3) NAVWF:CB (U/mL) -9 (-17 to 5)||§ -8 (-16 to 3) NAFXIIa (ng/mL) 0 (-25 to 43) -20 (-43 to 33) -14 (-43 to 0)Prothrombin F1 + 2 (nM) -19 (-55 to 32)¶ 91 (36 to 160) 27 (-12 to 180)‡C3a des arg (ng/mL) 0 (-41 to 65) -10 (-45 to 155) 16 (-27 to 295)

* WBC reduction was performed with an integral PLAS 4 WBC-reduction filter in the Maco Pharma MB pack configuration.† p < 0.01 refers to significance from the Mann-Whitney U-test between fresh and frozen plasma.‡ p < 0.05.§ n = 11.¶ p < 0.01 refers to significance from the Wilcoxon rank sum test between plasma before freezing and after thawing.|| p < 0.05.

27

METHYLENE BLUE PHOTOINACTIVATION OF PLASMA

Volume 43, September 2003 TRANSFUSION 1243

1.70 [1.24-2.31 g/L] non-WBC reduced) in plasma subse-quently MB treated and removed using the MB1 filter.VWF cleaving protease (VWF:CP) activity was measured infour MB-inactivated plasmas, which were WBC reducedbefore freezing and MB-removed using the MB1 removalfilter. VWF:CP results ranged from 0.81 to 1.00 (normalrange, 0.80-1.20 in citrated plasma). We have not assessedVWF:CP activity in plasma MB-depleted using the Blueflexfilter. Quality-monitoring data from routinely processed

units (n = 225), WBC-reduced beforefreezing, MB-treated, and MB-removedusing the MB1 filters, showed a meanvolume of 242 mL (SD = 19) and FVIIIcontent of 0.79 IU per mL (SD = 0.25).These results comply with UK specifica-tions for MB-treated FFP.20

Effect on cellular content of plasma

When platelets were spiked into plasma,there was no consistent differencebetween levels measured before andafter freeze-thawing when measured byhematology analyzer, but levels wereconsistently lower after thawing whenmeasured by flow cytometry (Table 4).When WBCs were spiked into plasma,there appeared to be a trend for lowerWBC counts and higher levels of a1-proteinase inhibitor: PMN elastasecomplexes in plasma after freezing(Table 4). However, in the absence ofplatelets, levels of LDH did not increasesubstantially. Freeze-thawing of plasma

resulted in an increase in levels of platelet microparticlesas well as microparticles characterized by the binding ofpurified annexin V (Table 4). Platelet microparticles werereduced to levels observed in fresh WBC-reduced plasmaor below after WBC reduction of frozen plasma with thePLAS 4 filter. However, a significant proportion (~30%) ofmicroparticles characterized by annexin V binding werenot removed by the WBC-reduction or the MB1 removalstep. These also appeared to be derived solely from plate-

TABLE 2. Effect of MB-removal filters on plasma coagulation activity using frozen-thawed MB photoinactivated plasma

FactorMB1 filter median (range) Blueflex filter median (range)

Before MB removal After MB removal Before MB removal After MB removalNumber 11 11PT (ratio) 1.17 (1.11-1.34) 1.07 (1.03-1.24)* 1.23 (1.13-1.33) 1.22 (1.13-1.32)*APTT (ratio) 1.11 (1.05-1.28) 1.13 (1.06-1.30)† 1.25 (1.17-1.39) 1.27 (1.19-1.43)*Fibrinogen (g/L) 1.88 (1.30-2.13) 1.93 (1.28-2.27)* 2.04 (1.37-2.13) 1.96 (1.28-2.32)FII (IU/mL) 0.96 (0.77-1.04) 1.00 (0.78-1.04)† 0.95 (0.77-1.04) 0.97 (0.77-1.04)FV (U/mL) 0.78 (0.55-0.86) 0.80 (0.56-0.91)* 0.76 (0.58-0.97) 0.76 (0.58-0.88)FVII (IU/mL) 0.99 (0.79-1.43) 1.02 (0.83-1.55)* 1.01 (0.77-1.40) 1.00 (0.78-1.45)FVIII (IU/mL) 0.63 (0.47-0.89) 0.62 (0.48-0.86) 0.61 (0.48-0.79) 0.61 (0.46-0.76)F IX (IU/mL) 0.96 (0.78-1.06) 0.96 (0.83-1.11)FX (IU/mL) 0.95 (0.70-1.12) 1.02 (0.75-1.15)* 0.95 (0.72-1.09) 0.96 (0.73-1.07)FXI (U/mL) 0.77 (0.57-0.99) 0.77 (0.59-0.99) 0.75 (0.55-0.99) 0.71 (0.58-0.78)FXII (U/mL) 0.88 (0.41-1.07) 0.88 (0.40-1.09)VWF:Ag (IU/mL) 0.99 (0.74-1.19) 0.98 (0.75-1.18) 0.92 (0.70-1.09) 0.93 (0.70-1.09)VWF:CB (U/mL) 0.60 (0.47-0.68) 0.73 (0.42-0.85) 0.70 (0.59-0.87) 0.74 (0.56-0.82)FXIIa (ng/mL) 1.50 (0.40-2.50) 1.80 (0.70-2.90)* 1.25 (0.75-2.00) 1.50 (1.00-2.25)*Prothrombin F1 + 2 (nM) 0.96 (0.61-1.75) 0.88 (0.76-1.83) 0.74 (0.58-1.19) 0.75 (0.52-1.25)C3a des arg (ng/mL) 438 (207-1928) 304 (107-431)* 337 (225-1909) 302 (226-1713)

* p < 0.01 refers to statistical significance from the Wilcoxon rank sum test between plasma before and after MB removal.† p < 0.05.

TABLE 3. Final levels of coagulation factors and activation markers in freeze-thawed MB-photoinactivated plasma after MB removal

Factor Final level in plasma* Reference range† Units in range (%)Number 22 66PT (ratio) 1.16 (1.03-1.32) 1.05 (0.95-1.16)‡ 50APTT (ratio) 1.24 (1.06-1.43) 1.09 (0.86-1.36)‡ 95Fibrinogen (g/L) 1.95 (1.28-2.32) 1.10-4.30 100FII (IU/mL) 0.98 (0.77-1.04) 0.70-1.20 100FV (U/mL) 0.79 (0.56-0.91) 0.50-1.40 100FVII (IU/mL) 1.01 (0.78-1.55) 0.60-1.40 100FVIII (IU/mL) 0.62 (0.46-0.86) 0.40-1.60 100F IX (IU/mL) 0.96 (0.83-1.11)§ 0.60-1.40 100FX (IU/mL) 1.01 (0.73-1.15) 0.70-1.30 100FXI (U/mL) 0.75 (0.58-0.99) 0.60-1.30 91FXII (U/mL) 0.88 (0.40-1.09)§ 0.40-1.50 100VWF:Ag (IU/mL) 0.96 (0.70-1.18) 0.60-1.65 100VWF:CB (U/mL) 0.74 (0.42-0.85) 0.50-1.50 91FXIIa (ng/mL) 1.50 (0.40-2.90) 0.50-5.00 100Prothrombin F1 + 2 (nM) 0.85 (0.52-1.83) 0.20-1.10 77C3a des arg (ng/mL) 303 (107-1713) 1117 (0-12,330) 100

* Data (n = 22) are represented by the median (range) from plasma MB-removed by MB1 filter (n = 11) and Blueflex filter (n = 11).

† Reference range of normal plasmas is defined as the mean ± 2 SD for normally distributed data and the geometric mean with 95-percent CI for skewed data based on WBC-reduced FFP. Percentage of units in range is defined as above the lower limit for coagulation factors and below the upper limit for PT, APTT, and activation markers.

‡ n = 100.§ n = 11, using MB1 filter only.

28

GARWOOD ET AL.

1244 TRANSFUSION Volume 43, September 2003

lets because levels after WBC reduction in samples spikedwith WBCs in the absence of platelets were not differentfrom WBC-reduced plasma alone. However, even at anadded platelet count of 30 ¥ 109 per L (Sample D, the cur-

rent UK specification), the number of annexin V-positivemicroparticles in frozen-thawed plasma subsequentlyfiltered using the PLAS 4 filter (43 ¥ 109/L) is similar tothat seen in our current routine WBC-reduced non-MB-treated plasma product (mean residual platelet count of 3¥ 109/L).

Before freezing, levels of RBC microparticles were 8(5-12 ¥ 109/L), increasing by 33 percent after freeze-thawing. This was not related to platelet or WBC content,and levels after WBC reduction were below the detectionof the assay system used (data not shown).

DISCUSSION

MB treatment of plasma has been shown to inactivate 4 to6 logs of transfusion-transmitted viruses, including HIV,HBV, parvovirus B19, and West Nile virus.5,21,22 Originalstudies on MB inactivation were reported on plasmafreeze-thawed before treatment.5,7 Later, work on othersystems (Baxter Pathinact and Maco Pharma Theraflexsystems) was performed on fresh plasma.10-12 However,there are no data available on the difference betweenusing fresh or freeze-thawed plasma as a starting compo-nent for MB treatment. In our study, freeze-thawing ofplasma resulted in a small loss of FXII and VWF:CB activity

Fig. 2. PT ratio in MB-treated (n ==== 22) or reference plasma (n ====

100). For MB-treated plasma, MB was removed by MB1 filter

(n ==== 11) and Blueflex filter (n ==== 11). Reference plasma is histor-

ical data from WBC-reduced FFP. Horizontal bar represents the

median value. The PT is expressed as a ratio to the geometric

mean result of 20 normal citrated plasmas.

0.80

1.00

1.20

1.40

PT

rat

io

Referenceplasma

MB-treatedplasma

TABLE 4. The effect of freeze-thawing, WBC reduction, and MB removal on cellular components of plasmaSpike*

A B C D E F G H I JPlatelets – HA† (109/L)

Before freeze <3 <3 10 31 108 <3 <3 <3 <3 <3After thaw <3 6 10 32 97 <3 <3 <3 <3 <3

Platelets – FC‡ (109/L)Before freeze <0.5 5.9 10.8 32.0 125.1 <0.5 <0.5 <0.5 <0.5 <0.5After thaw <0.5 4.8 9.0 24.6 92.2 <0.5 <0.5 <0.5 <0.5 <0.5

WBCs (106/L)Before freeze <1 8.9 29.6 90.7 233.8 <1 1.8 14.1 50.6 92.9After thaw <1 9.1 28.1 94.0 210.9 <1 1.6 13.2 43.4 89.5

PMN elastase (mg/L)Before freeze 46.4 39.7 37.5 46.3 57.0 20.6 23.7 24.0 25.3 25.7After thaw 37.5 35.8 42.2 53.5 79.8 22.8 23.1 30.4 43.2 58.8After WBC reduction 34.5 35.3 40.8 57.4 77.4 21.1 20.5 27.0 41.2 53.5After MB1 filter 35.1 37.0 45.1 58.2 71.9 18.1 21.3 24.8 37.9 57.9

Platelet microparticles (109/L)Before freeze 0.2 0.2 0.4 0.5 1.0 0.3 0.5 0.4 0.3 0.9After thaw 0.1 0.5 0.7 2.6 10.1 0.1 0.1 0.0 0.3 0.0After WBC reduction 0.0 0 0 0.1 0.2 0 0 0 0.1 0After MB1 filter 0.1 0.0 0.1 0.1 0.2 0.1 0.0 0.1 0.0 0.3

Annexin V +ve microparticles (109/L)Before freeze 6 7 8 10 13 4 4 4 5 4After thaw 12 31 54 138 574 9 9 9 13 13After WBC reduction 3 10 20 46 198 5 2 3 5 5After MB1 filter 3 8 15 43 196 4 4 4 4 8

LDH (U/mL)Before freeze 393 400 423 476 732 390 381 388 376 382After thaw 384 404 434 523 796 388 386 385 393 395

* WBC-reduced plasma units were spiked with WBCs alone (Samples G-J) or WBCs and platelets (Samples B-E) to the concentrations shown in the before freeze rows. Samples A and F were not spiked. Plasma was frozen-thawed, WBC reduced using the PLAS 4 filer; MB added and MB removed using the MB1 filter. Results are from a single experiment. Platelets and WBCs were not detectable following WBC reduction.

† FC-flow cytometry.‡ HA-haematology analyser.

29

METHYLENE BLUE PHOTOINACTIVATION OF PLASMA

Volume 43, September 2003 TRANSFUSION 1245

as well as a minor prolongation of PT and APTT. Interest-ingly, there was an apparent increase in activity of FVIIand FXI on freeze-thawing, presumably reflecting smallchanges in the activation status of these factors. However,other factors studied remained unchanged. This agreeswith the work of Zeiler et al.,7 who showed that the loss ofcoagulation activity in MB-treated plasma was mainlyattributable to the MB photoinactivation step rather thanfreeze-thawing of plasma. The loss of activity observed infrozen-thawed units due to MB photoinactivation in ourstudy was similar to that previously reported.5,7 For thevariables we studied, with the exception of FVII, loss ofcoagulation factor activity due to the MB-inactivation step(including the WBC reduction filter) was 8-percent higherwhen frozen-thawed plasma units were used rather thanfresh. In addition, the increase in prothrombin F1 + 2 lev-els after MB inactivation and WBC reduction was higherin units frozen-thawed compared with fresh plasma, indi-cating a higher degree of thrombin generation. This wasnot associated with an increase in FXIIa. Because our MBprocess includes an integral WBC-reduction step, we can-not determine whether the differences seen between freshand frozen-thawed plasma are attributable to the WBC-reduction or MB process. The WBC-reduction filter usedin the MB packs has previously been shown to have min-imal effect on coagulation activity (unpublished data)using fresh plasma, but this could be different for frozen-thawed plasma.

We also sought to compare the effect of two differenttypes of MB removal filters (Maco Pharma Blueflex or PallMB1 filter) on plasma factor activity. Neither filter resultedin loss of any variable studied. Both MB removal filtersresulted in a small increase in the APTT ratio, which mightbe attributable to contact activation of plasma with thefilter. This is difficult to assess because although both fil-ters increased FXIIa antigen, this assay is influenced byMB. However, levels of FXIIa antigen in the final MB-removed component were not higher than untreatedplasma units. The filtration times for one filter (MacoPharma Blueflex) were longer compared with the other(Pall MB1) (5-11 vs. 2-5 min, respectively), but the loss ofplasma was equivalent for both filters. However, there didnot appear to be any difference between the two filters interms of activation of the contact system or thrombin gen-eration as evidenced by the generation of FXIIa or pro-thrombin F1 + 2. Our results using the MB1 filter comparewell with that previously published,14 showing minimalloss of clotting factor activity. However, there was anapparent increase in levels of fibrinogen and PT-derivedcoagulation factors after filtration with the MB1 filter,which was associated with a small decrease in the PT ratio.We cannot explain these results because these assays donot appear to be influenced by the presence of MB, butthey could possibly be a result of small increases in theactivation state of coagulation factors. The changes in

coagulation factors observed after filtration with eitherMB-removal filter appear to be clinically insignificant.Neither MB-removal filter resulted in generation of C3ades arg, a marker of complement activation. However, lev-els after filtration were reduced using the Pall MB1 filter.Whether this has any clinical benefit in terms of acutereactions is unclear.

As well as examining the loss of coagulation factorsduring MB photoinactivation, we compared residual lev-els in the final component with reference ranges based onprevious studies of nontreated FFP in our laboratories.Despite the observed losses of coagulation factors due toMB treatment, final levels of all coagulation factors infrozen-thawed, MB-treated, and removed FFP were abovethe lower limit of the reference range in over 90 percent ofunits. However, over 50 percent of units were above theupper reference range for PT ratio. This presumablyreflects the loss of fibrinogen and FII, FV, FVII, and FXbecause the PT is dependent upon these factors. However,the PT ratio of all units was less than 1.35. In addition, 23percent of units had levels of prothrombin F1 + 2 higherthan the upper limit of the reference range, reflecting theincrease in prothrombin F1 + 2 seen due to the MB pro-cess. The clinical significance of increased F1 + 2 levels isunclear, but values higher than those observed in ourstudy are seen in S/D-treated plasma.23 We did not assessthe effect of MB treatment or removal on plasma inhibi-tors of coagulation. However, MB treatment is reportedto have minimal effect on levels of antithrombin, a2-antiplasmin and protein C & S.5,7,11 After MB removal (PallMB1 filter), levels of antithrombin and protein C & S arewithin the normal range.14

We also evaluated the addition of a WBC-reductionstep before freezing plasma, which did not appear to aug-ment the loss of fibrinogen and FVIII activity. VWF:CPactivity in plasma that was WBC reduced, MB inactivated,and removed using the MB1 filter was also within refer-ence ranges established by other laboratories, suggestingthat these processing steps do not have a major effect onVWF:CP using the techniques employed. These results areconsistent with our previous findings on MB-treated freshplasma24 and others results on frozen-thawed plasma.25

However, we cannot exclude small losses of activity giventhe relatively small number of samples used in this study.

In the UK, all blood components are WBC-reducedbefore storage. However, for logistical reasons we alsowanted the flexibility to freeze non-WBC-reduced plasmaintended for MB treatment, as a subsequent WBC-reduction step is integral to the Maco Pharma MB process.We were concerned that freeze-thawing may result inreduced cell removal by the WBC-reduction filter or causefragmentation of cells with the potential for increasing therisk of transmission of vCJD. Most cellular prion protein inblood, used here as a surrogate marker for the potentiallyinfective abnormal prion protein, is associated with

30

GARWOOD ET AL.

1246 TRANSFUSION Volume 43, September 2003

plasma (65%), with 26.5 percent, 1.8 percent, and 0.8 per-cent associated with platelets, RBCs, and polymorphonu-clear cells, respectively.13 Our results indicate that themajority of WBCs can be detected after thawing, and theseare removed to undetectable levels after the WBC-reduction step of the MB process. However, the methodwe employed to detect WBCs predominantly measuresWBC nuclei (unpublished data), and therefore provideslittle information on cellular integrity. The majority ofWBCs in freeze-thawed plasma are detectable with PIwithout prior permeabilization,26,27 suggesting that freeze-thawing alters WBC membrane integrity. The increase inlevels of a1-proteinase inhibitor: PMN elastase complexesafter thawing of plasma spiked with WBCs shows thatPMN degranulation is occurring, but the postthaw levelsremain below 100 mg per L, which is not suggestive oflarge-scale PMN disintegration. Furthermore, in theabsence of platelets, levels of LDH did not increase sub-stantially after freezing, suggesting that WBCs do not dis-integrate. We were unable to assess WBC fragments due tothe insufficient sensitivity of available methods.

When platelets were spiked into plasma, there was anincrease in platelet-derived microparticles after freeze-thawing of plasma, which probably explains the smalldecrease in platelet count detected by flow cytometrybecause these events would not be included in the plateletcount. This fall was not detected by hematology analyzer,possibly because cell fragments can be detected as plate-lets by impedance-based methods. These fragments werereduced to or below the level in WBC-reduced freshplasma after the WBC-reduction step of the MB process.However, we also analyzed cell microparticles based onthe binding of purifed annexin V, which has a high affinityfor anionic phospholipids. Freeze-thawing resulted inan increase in annexin-V-positive microparticles, whichappear to be mainly derived from platelets and were onlypartially removed by WBC reduction. The increaseddetection of microparticles by this method compared withusing an antibody against the platelet receptor CD61 isprobably attributable to the greater number of moleculesper platelet of anionic phospholipid (1 ¥ 106) comparedwith CD61 (4-8 ¥ 104).28,29 The presence of RBC and WBCmicroparticles (which will also bind annexin V) may alsohelp to explain this difference, but this seems unlikelybecause in the absence of platelets the differencesbetween methods were less pronounced. The number ofannexin-V-positive microparticles found in non-WBC-reduced plasma that has been frozen-thawed and thenfiltered is not appreciably higher than would be found inplasma that we currently produce.

The effect of loss of coagulation factor activity due toMB treatment on the in vivo efficacy of the component isdifficult to assess because there are no published random-ized, controlled clinical trial data comparing MB to eitherstandard FFP or S/D-treated FFP. However, 2.5 million

units of MB FFP have been transfused internationallywithout obvious clinical sequelae.30,31 In Spain, the switchfrom standard to MB-treated FFP has been associatedwith an increase in demand for FFP and cryoprecipitate,32

which the authors attribute to loss of coagulation activity.However the increase in use (56%) appears to be dispro-portionate to the decrease in coagulation factors, suggest-ing that other factors, such as perception of a safercomponent, may have been influencing usage. It is alsoreported that the use of MB FFP is associated with a highernumber of plasma exchanges compared with untreatedFFP for the treatment of thrombotic thrombocytopenicpurpura,33 although we found no difference in the levelsof VWF cleaving activity, the presumed therapeutic moi-ety in plasma treatment of thrombotic thrombocytopenicpurpura, in MB FFP.24 It is critical that transfusion servicesintroducing pathogen inactivation of components moni-tor ongoing trends in usage as well as having a system forhazard reporting. At the time of writing, MB-treated and-removed FFP is routinely produced in England and Walesfor transfusion to children and neonates born after 1995,with similar arrangements in other parts of the UK. How-ever, in the near future, plasma to be pathogen inactivatedfor this patient group throughout the UK will be importedfrom volunteer donors in North America. Processes cur-rently available for the pathogen inactivation of plasma allresult in a decrease in coagulation factor activity. Improve-ments in the safety of blood need to be balanced againstsome likely reduction in the component potency. Single-unit systems for pathogen inactivation of plasma thathave less effect on coagulation factor activity are clearlydesirable.

ACKNOWLEDGMENTS

We are grateful to the Haemostasis Research Unit (University

College London, UK) for performing VWF:CP assays and Saber

Bashir, PhD, (National Blood Service, Brentwood, UK) for per-

forming LDH assays.

REFERENCES

1. Serious Hazards of Transfusion Report 2000-2001. Published

2002 by Serious Hazards of Transfusion Steering Committee,

Manchester, UK.

2. Brown P, Rohwer RG, Dunstan BC, et al. The distribution of

infectivity in blood components and plasma derivatives in

experimental models of transmissible spongiform

encephalopathy. Transfusion 1998;38:810-6.

3. Brown P, Cervenakova L, McShane LM, et al. Further studies

of blood infectivity in an experimental model of

transmissible spongiform encephalopathy, with an

explanation of why blood components do not transmit

Creutzfeldt-Jakob disease in humans. Transfusion 1999;

39:1169-78.

31

METHYLENE BLUE PHOTOINACTIVATION OF PLASMA

Volume 43, September 2003 TRANSFUSION 1247

4. Hunter N, Foster J, Chong A, et al. Transmission of prion

diseases by blood transfusion. J Gen Virol 2002;83:

2897-905.

5. Lambrecht B, Mohr H, Knüver-Hopf J, Schmitt H.

Photoinactivation of viruses in human plasma by

phenothiazine dyes in combination with visible light. Vox

Sang 1991;60:207-13.

6. Inada Y, Hessel B, Blomback B. Photooxidation of fibrinogen

in the presence of methylene blue and its effect on

polymerization. Biochim Biophys Acta 1978;532:161-70.

7. Zeiler T, Riess H, Wittmann G, et al. The effects of methylene

blue phototreatment on plasma proteins and in vitro

coagulation capability of single-donor fresh frozen plasma.

Transfusion 1994;34:685-9.

8. Aznar JA, Molina R, Montoro JM. Factor VIII/von Willebrand

factor complex in methylene blue-treated fresh plasma.

Transfusion 1999;39:748-50.

9. Aznar JA, Bonanad S, Montoro JM, et al. Influence of

methylene blue photoinactivation treatment on coagulation

factors from fresh frozen plasma, cryoprecipitates and

cryosupernatants. Vox Sang 2000;79:156-60.

10. Hornsey VS, Krailadsiri P, MacDonald S, et al. Coagulation

factor content of cryoprecipitate prepared from methylene

blue plus light virus inactivated plasma. Br J Haematol

2000;109:665-70.

11. Hornsey VS, Drummond O, Young D, et al. A potentially

improved approach to methylene blue inactivation of

plasma: the Maco Pharma Maco-Tronic system. Transfus

Med 2001;11:31-6.

12. Seghatchian J, Krailadsiri P. What’s happening: the quality of

methylene blue treated FFP and cryo. Trans Apheresis Sci

2001;25:227-31.

13. MacGregor I, Hope J, Barnard G, et al. Application of a time

resolved fluoroimmunoassay for the analysis of normal

prion protein in human blood and its components. Vox Sang

1999;77:88-96.

14. Riggert J, Humpe A, Legler TJ, et al. Filtration of methylene

blue-photooxidized plasma: influence on coagulation and

cellular contamination. Transfusion 2001;41:82-6.

15. AuBuchon JP, Pickard C, Herschel L, et al. Removal of

methylene blue from plasma via an adsorbent filter. Vox

Sang 1998;74:1-6.

16. Cardigan R, Sutherland J, Garwood M, et al. The effect of

leucocyte depletion on the quality of fresh frozen plasma. Br

J Haematol 2001;114:233-40.

17. Allford SL, Harrison P, Lawrie AS, et al. von Willebrand

factor-cleaving protease activity in congenital thrombotic

thrombocytopenic purpura. Br J Haematol 2000;111:1215-

22.

18. Krailadsiri P, Seghatchian J, Williamson LM. Platelet storage

lesion of WBC-reduced, pooled, buffy coat-derived platelet

concentrates prepared in three in-process filter/storage bag

combinations. Transfusion 2001;41:243-50.

19. Verpoort T, Chollet S, Lebrun F, Goudaliez F, Mohr H,

Walker WH. Elimination of methylene blue from photo-

dynamically treated virus inactivated fresh-frozen plasma:

the Blueflex filter. Trans Clinique Biologique 2001;8 (Suppl

1):103s.

20. Guidelines for the blood transfusion services in the United

Kingdom (2002), 6th ed. London: The Stationary Office, 2002.

21. Mohr H, Kneuver-Hopf J. Influence of photodynamic

treatment on parvovirus B19. Infusionstherapie

Transfusionsmedizin 1999;26 (Suppl 1):1.18.

22. Mohr H, Lambrecht B, Selz A. Photodynamic virus

inactivation of blood components. Immunol Invest

1995;24:73-85.

23. Hellstern P, Sachse H, Schwinn H, Oberfrank K.

Manufacture and in vivo characterization of a solvent/

detergent-treated human plasma. Vox Sang 1992;63:178-85.

24. Cardigan R, Allford S, Williamson L. Levels of von Willebrand

factor cleaving protease are normal in methylene blue

treated fresh frozen plasma. Br J Haematol 2002;117:253-4.

25. Furlan M, Robles R, Galbusera M, et al. von Willebrand

factor-cleaving protease in thrombotic thrombocytopenic

purpura and the hemolytic-uremic syndrome. N Engl J Med

1998;339:1578-84.

26. Rider JR, Winter MA, Payrat JM, et al. Leucocytes can be

eliminated from plasma by filtration prior to viral

inactivation with methylene blue. Vox Sang 1998;74:209-10.

27. Hiruma K, Okuyama Y. Effect of leucocyte reduction on the

potential alloimmunogenicity of leucocytes in fresh-frozen

plasma products. Vox Sang 2001;80:51-6.

28. Wagner CL, Mascelli MA, Neblock DS, et al. Analysis of

GPIIb/IIIa receptor number by quantification of 7E3 binding

to human platelets. Blood 1996;88:907-14.

29. Tait JF, Smith C. Wood BL. Measurement of phosphatidyl-

serine exposure in leukocytes and platelets by whole-blood

flow cytometry with annexin V. Blood Cells Mol Dis 1999;25:

271-8.

30. Mohr H. Methylene blue and thionine in pathogen

inactivation of plasma and platelet concentrates. Trans

Apheresis Sci 2001;25:183-4.

31. Prowse C, Robinson AE. Pathogen inactivation of labile

blood components. Transfus Med 2001;11:147.

32. Atance R, Pereira A, Ramírez B. Transfusing methylene blue-

photoinactivated plasma instead of FFP is associated with

an increased demand for plasma and cryoprecipitate.

Transfusion 2001;41:1548-52.

33. de la Rubia J, Arriaga F, Linares D, et al. Role of methylene

blue-treated or fresh-frozen plasma in the response

to plasma exchange in patients with thrombotic thrombocy-

topenic purpura. Br J Haematol 2001;114:721-3.

32

04590459

F - 59420 MOUVAUX

Specifications

. Filters : Plasmaflex PLAS4, Blueflex filter

. Bags : 2 PVC

. Included Items : Methylene Blue pill (85µg)

Use

. Label : English, French, German, Dutch

. Sterilisation : Steam

. Shelf life : 2 years

. Packaging : 2 packs/peelable sachet - 24 packs/box

Ref. SDV0001XQ

Pathogen Reduction SystemTHERAFLEX - MB PLASMA

SD

V00

01X

Q_A

_A08

/03/

2006

Pathogen Reduction ofLeucodepleted PlasmaMethylene BlueRemoval by Filtration

1/

2/

3/ 4/

5/

Illumination Filtration

1

2

Sterile Connection

33